{ "edition": "2026-11-07", "name": "Rubric, edition of Saturday 7 November 2026", "url": "https://rubric.hrih.si/editions/2026-11-07/", "published_at": "2026-11-07T06:00:00Z", "rubric_version": "2.3", "how_ranking_works": "https://rubric.hrih.si/how-ranking-works/", "license": { "name": "CC BY 4.0", "url": "https://creativecommons.org/licenses/by/4.0/", "attribution": "Rubric (rubric.hrih.si)", "notice": "CC BY 4.0, Rubric (rubric.hrih.si), edition of 2026-11-07: https://rubric.hrih.si/editions/2026-11-07/" }, "models": { "jev": "jev-1.14.0", "judge_a": "claude-haiku-5-5", "judge_b": "gpt-5.6-luna", "writer": "claude-sonnet-5-5", "claim_check": "claude-sonnet-5-5 + jev-1.14.0", "summarizer": "claude-haiku-5-5" }, "fields": { "rank": "Place in the edition; since rubric 2.3 after the placement terms (see score_rank), before it the place by score", "id": "The story id in Rubric", "title": "The title as shown", "url": "The source: the original article, paper or release", "domain": "The host name of the source", "site": "The site of the source, as Rubric names it", "categories": "Topic keys; \"other\" when the story fits none", "category_names": "The topics as the site names them", "format": "The kind of piece: paper, write-up, release, analysis, tutorial, news, new project", "depth": "How far the story was checked: claim_checked, three_judges or two_judges", "published_at": "The publication time of the source, when known", "jev": "Judge 1 (jev-1.14.0): its own weighted sum of the rubric questions, 0 to 1, before penalties; null when it did not score the story", "judge_a": "Judge 2 (claude-haiku-5-5): its own weighted sum of the rubric questions, 0 to 1, before penalties; null when it did not score the story", "judge_b": "Judge 3 (gpt-5.6-luna): its own weighted sum of the rubric questions, 0 to 1, before penalties; null when it did not score the story", "score": "The published total: the mean answer of the panel per question, weighted, minus the claim-check and freshness penalties", "contested": "True when the judges disagreed on at least one question by more than the contested threshold", "flag_penalty": "Taken off the total for claims the checkers both flagged", "freshness_penalty": "Taken off the total for older work", "score_rank": "Rubric 2.3+: the place by score alone; null in editions ranked before it", "topic_penalty": "Rubric 2.3+: taken off for placement only, for stories of the same topic placed above it beyond the free ones; not part of score", "repeat_penalty": "Rubric 2.3+: taken off for placement only, for similarity to a story placed above it today or ranked in a recent edition; not part of score", "summary": "The summary of the source by Rubric, written by AI", "page": "The page of the story on Rubric, with the full breakdown", "card": "The social card of the story (PNG, 1200 × 630)", "breakdown": "JSON only: the weight of each question and the answer of each judge, with the mean and spread of the panel" }, "stories": [ { "rank": 1, "id": "st_20261107_16888", "title": "Compromised Certificates and OpenJDK Java Runtimes", "url": "https://openjdk.org/emudhra", "domain": "openjdk.org", "site": "openjdk.org", "categories": [ "programming", "security", "it" ], "category_names": [ "Programming & dev tools", "Security", "IT & networking" ], "format": "Tutorial", "depth": "claim_checked", "published_at": "2026-11-06T14:15:55.000Z", "jev": 0.7699, "judge_a": 0.7246, "judge_b": 0.7582, "score": 0.7508, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 1, "topic_penalty": 0, "repeat_penalty": 0, "summary": "OpenJDK published guidance after the private keys of eight eMudhra intermediate CA certificates were reported compromised. Standalone runtimes should be updated to a release citing JDK-8395690, or else have four root certificates (three emSign roots and comodoaaaca) deleted from the cacerts keystore, or have revocation checking enabled. A CheckCerts.java tool tests exposure. Most Linux packages follow OS certificate updates, while Debian-based systems need the keystore rebuilt by hand because of a known bug. FreeBSD and Homebrew need package updates. Several of these distribution details were not independently verified.", "page": "https://rubric.hrih.si/editions/2026-11-07/1/", "card": "https://rubric.hrih.si/og/2026-11-07/1.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.728, "judge_a": 0.62, "judge_b": 0.65, "mean": 0.6658, "spread": 0.108 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.76, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8133, "spread": 0.12 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.72, "judge_a": 0.67, "judge_b": 0.6, "mean": 0.6633, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.973, "judge_a": 0.92, "judge_b": 0.9, "mean": 0.9308, "spread": 0.073 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.61, "judge_a": 0.72, "judge_b": 0.8, "mean": 0.71, "spread": 0.19 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.963, "judge_a": 0.8, "judge_b": 0.68, "mean": 0.8142, "spread": 0.283 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.79, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.7033, "spread": 0.19 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.03, "judge_b": 0.02, "mean": 0.0333, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.17, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.15, "spread": 0.12 } } }, { "rank": 2, "id": "st_20261107_16380", "title": "An Android app with no code or permissions can trap phones in a boot loop", "url": "https://www.bitdefender.com/en-us/blog/labs/ghostmap-android", "domain": "bitdefender.com", "site": "Bitdefender Labs", "categories": [ "security" ], "category_names": [ "Security" ], "format": "Paper", "depth": "claim_checked", "published_at": "2026-11-06T00:58:50.000Z", "jev": 0.7009, "judge_a": 0.7208, "judge_b": 0.7928, "score": 0.7381, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 2, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Bitdefender Labs reports \"GhostMap,\" an Android flaw in which an app with no code or permissions, never opened, can send system_server into a boot loop that safe mode doesn't stop. The cause is a mismatch between libziparchive and libandroidfw in handling ZIP entries with unusual compression methods. Bitdefender reproduced it on a Pixel 10 with Android 17 and says it affects Android 10 and later. Google said it won't fix it through a security update. Bitdefender has seen no exploitation, and the claims come from its own testing.", "page": "https://rubric.hrih.si/editions/2026-11-07/2/", "card": "https://rubric.hrih.si/og/2026-11-07/2.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.673, "judge_a": 0.72, "judge_b": 0.82, "mean": 0.7375, "spread": 0.147 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.9, "judge_b": 0.93, "mean": 0.8533, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.663, "judge_a": 0.8, "judge_b": 0.72, "mean": 0.7278, "spread": 0.137 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.713, "judge_a": 0.7, "judge_b": 0.79, "mean": 0.7342, "spread": 0.09 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.973, "judge_a": 0.85, "judge_b": 0.92, "mean": 0.9144, "spread": 0.123 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.9525, "spread": 0.098 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9267, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.2, "judge_b": 0.18, "mean": 0.1433, "spread": 0.15 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.25, "judge_a": 0.35, "judge_b": 0.2, "mean": 0.2667, "spread": 0.15 } } }, { "rank": 3, "id": "st_20261107_17249", "title": "Musl – OpenSSH password auth bypass analysis", "url": "https://www.openwall.com/lists/musl/2026/11/06/7", "domain": "openwall.com", "site": "openwall.com", "categories": [ "security", "it" ], "category_names": [ "Security", "IT & networking" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T19:13:59.000Z", "jev": 0.6718, "judge_a": 0.7037, "judge_b": 0.8365, "score": 0.7373, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 3, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Rich Felker, musl's maintainer, published a root-cause analysis of an OpenSSH password bypass affecting versions before 10.6p2 on musl when UsePAM=no and PasswordAuthentication=yes. OpenSSH uses another user's hash as salt to mask timing for locked accounts. An overlong password makes SHA256/SHA512 crypt fail, returning musl's \"*\" failure token, which matches a locked \"*\" entry. OpenSSH now rejects short hashes outright; musl may switch tokens. The claims are the maintainer's own and not independently verified.", "page": "https://rubric.hrih.si/editions/2026-11-07/3/", "card": "https://rubric.hrih.si/og/2026-11-07/3.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.605, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.645, "spread": 0.23 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.72, "judge_a": 0.8, "judge_b": 0.93, "mean": 0.8167, "spread": 0.21 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.553, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.5911, "spread": 0.067 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.753, "judge_a": 0.8, "judge_b": 0.86, "mean": 0.8042, "spread": 0.108 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.803, "judge_a": 0.85, "judge_b": 0.94, "mean": 0.8644, "spread": 0.137 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.99, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.95, "spread": 0.09 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.85, "judge_b": 0.98, "mean": 0.91, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.02, "judge_b": 0.01, "mean": 0.0367, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.15, "judge_b": 0.04, "mean": 0.1167, "spread": 0.12 } } }, { "rank": 4, "id": "st_20261107_17242", "title": "Restaurant Delivery Payment-Skimming Campaign", "url": "https://www.infohawk.com/news/2026-11-05-scamwich-restaurant-skimming", "domain": "infohawk.com", "site": "InfoHawk", "categories": [ "security", "fintech" ], "category_names": [ "Security", "Fintech" ], "format": "Paper", "depth": "claim_checked", "published_at": "2026-11-06T19:19:27.000Z", "jev": 0.6421, "judge_a": 0.7133, "judge_b": 0.7973, "score": 0.7175, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 4, "topic_penalty": 0.02, "repeat_penalty": 0, "summary": "InfoHawk researchers report 820 look-alike domains impersonating 35 restaurant chains, 808 of them on .delivery, with Jimmy John's, Firehouse Subs and Jersey Mike's most cloned. The sites reverse-proxy the real ordering pages and swap in a checkout that collects card details and the bank's SMS code, so no restaurant site needs to be breached. The figures come from InfoHawk's own two research passes, with no independent source given. For teams running ordering or payment flows, it shows why typosquat monitoring and one-time-code phishing defenses matter.", "page": "https://rubric.hrih.si/editions/2026-11-07/4/", "card": "https://rubric.hrih.si/og/2026-11-07/4.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.583, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6542, "spread": 0.198 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.71, "judge_a": 0.85, "judge_b": 0.94, "mean": 0.8333, "spread": 0.23 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.483, "judge_a": 0.7, "judge_b": 0.62, "mean": 0.6011, "spread": 0.217 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.67, "judge_a": 0.75, "judge_b": 0.76, "mean": 0.7267, "spread": 0.09 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.903, "judge_a": 0.8, "judge_b": 0.86, "mean": 0.8544, "spread": 0.103 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.933, "judge_a": 0.75, "judge_b": 0.83, "mean": 0.8375, "spread": 0.183 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.93, "judge_a": 0.85, "judge_b": 0.97, "mean": 0.9167, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.03, "judge_b": 0.01, "mean": 0.0367, "spread": 0.06 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.16, "spread": 0.08 } } }, { "rank": 5, "id": "st_20261107_17400", "title": "Developers say they didn't know gov't was buying user data", "url": "https://404privacy.com/blog/app-developers-say-they-didnt-know-user-data-went-to-the-government/", "domain": "404privacy.com", "site": "404 Privacy", "categories": [ "security", "policy" ], "category_names": [ "Security", "Policy & regulation" ], "format": "Analysis", "depth": "claim_checked", "published_at": "2026-11-06T21:43:39.000Z", "jev": 0.7031, "judge_a": 0.6793, "judge_b": 0.6892, "score": 0.6905, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 9, "topic_penalty": 0, "repeat_penalty": 0, "summary": "404 Privacy's Seth Honda argues that app developers cannot claim ignorance after Bloomberg and Citizen Lab reported that Penlink's Webloc tool sells governments searchable phone-location histories, reportedly covering 12 months, drawn from the ad ecosystem. Developers of AccuWeather and Muslim Pro said they had no relationship with Penlink. Honda says embedding a location-collecting ad SDK feeds data into a market that includes government buyers, and urges developers to audit SDKs and test network traffic. The piece is opinion, and most figures lack cited sources.", "page": "https://rubric.hrih.si/editions/2026-11-07/5/", "card": "https://rubric.hrih.si/og/2026-11-07/5.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.725, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.735, "spread": 0.08 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.69, "judge_a": 0.85, "judge_b": 0.88, "mean": 0.8067, "spread": 0.19 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.79, "judge_a": 0.8, "judge_b": 0.62, "mean": 0.7367, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.928, "judge_a": 0.7, "judge_b": 0.76, "mean": 0.7958, "spread": 0.228 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.427, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.5656, "spread": 0.293 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.703, "judge_a": 0.4, "judge_b": 0.44, "mean": 0.5142, "spread": 0.303 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.8, "judge_b": 0.97, "mean": 0.8933, "spread": 0.17 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.03, "judge_b": 0.01, "mean": 0.04, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.4, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5233, "spread": 0.32 } } }, { "rank": 6, "id": "st_20261107_16493", "title": "Shaping and Monitoring Models Through Their Internal Representations", "url": "https://www.goodfire.com/research/mechanistic-shaping", "domain": "goodfire.com", "site": "Goodfire", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "claim_checked", "published_at": "2026-11-06T04:59:52.000Z", "jev": 0.6414, "judge_a": 0.6383, "judge_b": 0.78, "score": 0.6865, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 11, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Goodfire's paper proposes \"mechanistic shaping,\" which adds a training penalty on a model's reliance on, or expression of, an internal concept direction. On chest X-rays, reliance shaping with agent-selected drain features beat balanced-data training. In DPO on OLMo-3-7B, it kept capability gains while sycophancy stayed near baseline, across 7B to 32B models. Expression shaping reportedly curbed Countdown reward hacking in Qwen3.5-9B, though those figures are unchecked. Probes sometimes broke after shaping, but retrained probes recovered the concept in every experiment.", "page": "https://rubric.hrih.si/editions/2026-11-07/6/", "card": "https://rubric.hrih.si/og/2026-11-07/6.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.7, "judge_a": 0.62, "judge_b": 0.82, "mean": 0.7133, "spread": 0.2 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.8, "judge_b": 0.94, "mean": 0.7967, "spread": 0.29 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.713, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6778, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.568, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.5825, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.917, "judge_a": 0.8, "judge_b": 0.86, "mean": 0.8589, "spread": 0.117 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.87, "judge_a": 0.85, "judge_b": 0.93, "mean": 0.8833, "spread": 0.08 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.85, "judge_b": 0.91, "mean": 0.8833, "spread": 0.06 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.05, "judge_b": 0.08, "mean": 0.07, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.2, "judge_b": 0.22, "mean": 0.2367, "spread": 0.09 } } }, { "rank": 7, "id": "st_20261107_17454", "title": "Libghostty 1.0", "url": "https://mitchellh.com/writing/libghostty-1-0", "domain": "mitchellh.com", "site": "Mitchell Hashimoto", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "Release", "depth": "claim_checked", "published_at": "2026-11-06T00:00:00.000Z", "jev": 0.7494, "judge_a": 0.6565, "judge_b": 0.6513, "score": 0.6857, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 12, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Mitchell Hashimoto has released libghostty-vt 1.0, the C terminal-emulation library extracted from Ghostty, with a stable C ABI enforced in CI through libabigail checks and versioned ELF symbols. It also ships signed tarballs for offline distro packaging, and it builds for Linux, macOS, Windows, Android, iOS and WebAssembly. JetBrains IDE terminals now use it, and a Neovim replacement of libvterm is approved but awaiting a tagged release. Claims of broader usage, the 10,000-terminal test and the 1,500-line terminal come from the author and are unverified.", "page": "https://rubric.hrih.si/editions/2026-11-07/7/", "card": "https://rubric.hrih.si/og/2026-11-07/7.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.735, "judge_a": 0.6, "judge_b": 0.65, "mean": 0.6617, "spread": 0.135 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.79, "judge_a": 0.85, "judge_b": 0.85, "mean": 0.83, "spread": 0.06 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.753, "judge_a": 0.6, "judge_b": 0.45, "mean": 0.6011, "spread": 0.303 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.87, "judge_a": 0.75, "judge_b": 0.7, "mean": 0.7733, "spread": 0.17 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.857, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.7589, "spread": 0.257 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.853, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7775, "spread": 0.153 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.92, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.15, "judge_b": 0.18, "mean": 0.14, "spread": 0.09 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.31, "judge_a": 0.3, "judge_b": 0.3, "mean": 0.3033, "spread": 0.01 } } }, { "rank": 8, "id": "st_20261107_17374", "title": "The Components of Agent Memory", "url": "https://www.anthropic.com/research/components-of-agent-memory", "domain": "anthropic.com", "site": "anthropic.com", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T21:19:24.000Z", "jev": 0.6532, "judge_a": 0.6046, "judge_b": 0.7925, "score": 0.6834, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 14, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Anthropic describes \"dreaming,\" a background pass where Claude reviews session transcripts to correct, reorganize and promote its memory notes. In the company's own tests, scaling dreaming from two to 128 sessions on a 350,000-file corpus raised wiki-based scores from 13% to 48%. On merger questions, the wiki scored 87% against 73% for search alone. Gains were uneven: single-document tasks showed none, and one wiki grew larger than its sources. Most figures come from Anthropic's internal evaluations and are not independently sourced.", "page": "https://rubric.hrih.si/editions/2026-11-07/8/", "card": "https://rubric.hrih.si/og/2026-11-07/8.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.68, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6533, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.8, "judge_b": 0.91, "mean": 0.7867, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.573, "judge_a": 0.67, "judge_b": 0.8, "mean": 0.6811, "spread": 0.227 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.725, "judge_a": 0.75, "judge_b": 0.83, "mean": 0.7683, "spread": 0.105 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.853, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.7711, "spread": 0.26 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.793, "judge_a": 0.7, "judge_b": 0.8, "mean": 0.7642, "spread": 0.1 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.8, "judge_b": 0.93, "mean": 0.87, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.1833, "spread": 0.13 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.15, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.19, "spread": 0.18 } } }, { "rank": 9, "id": "st_20261107_16936", "title": "Backslashes can accidentally allow users to access protected routes", "url": "https://octri.dev/blog/backslash-nginx-node-admin-route", "domain": "octri.dev", "site": "octri.dev", "categories": [ "programming", "security", "devops" ], "category_names": [ "Programming & dev tools", "Security", "Infrastructure & DevOps" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T14:41:22.000Z", "jev": 0.7607, "judge_a": 0.617, "judge_b": 0.7606, "score": 0.7127, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 5, "topic_penalty": 0.04, "repeat_penalty": 0, "summary": "In a write-up on octri.dev, the author shows an Nginx exact-match block on /admin/dashboard being bypassed with /admin%5Cdashboard: Nginx passed the normalized path, and Node's WHATWG URL parser treated the backslash as a separator and ran the admin handler. A 750-request test across 10 configurations also found routers disagreeing on encoded slashes. The test results are the author's own and unsourced. For builders, the point is to authorize the route the backend resolves, not rely on edge path rules.", "page": "https://rubric.hrih.si/editions/2026-11-07/9/", "card": "https://rubric.hrih.si/og/2026-11-07/9.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.643, "judge_a": 0.45, "judge_b": 0.68, "mean": 0.5908, "spread": 0.23 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.78, "judge_a": 0.8, "judge_b": 0.86, "mean": 0.8133, "spread": 0.08 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.813, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.6778, "spread": 0.213 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.915, "judge_a": 0.8, "judge_b": 0.82, "mean": 0.845, "spread": 0.115 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.86, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7467, "spread": 0.26 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.993, "judge_a": 0.8, "judge_b": 0.76, "mean": 0.8508, "spread": 0.233 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.5, "judge_b": 0.93, "mean": 0.7567, "spread": 0.43 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.0367, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.26, "judge_a": 0.35, "judge_b": 0.08, "mean": 0.23, "spread": 0.27 } } }, { "rank": 10, "id": "st_20261107_17011", "title": "Harness adapters: Safe RL for efficient agentic development", "url": "https://huggingface.co/blog/harness-hub/harness-adapters", "domain": "huggingface.co", "site": "huggingface.co", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T15:37:14.000Z", "jev": 0.646, "judge_a": 0.614, "judge_b": 0.7524, "score": 0.6708, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 18, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Hugging Face's harness-hub team released LoRA \"harness adapters\" for LFM2.5-24B-A2B, LFM3-1.7B and TwIL-8B, trained with GRPO inside unmodified pi, opencode and hermes harnesses. One vLLM server hosts all adapters, and a rule-based proxy, safeagents, blocks reward-hacking tool calls and scores them 0. The authors report fewer tool calls at equal or better held-out accuracy, but only from smoke tests, and most figures lack independent sourcing. The new task set is 94% build tasks.", "page": "https://rubric.hrih.si/editions/2026-11-07/10/", "card": "https://rubric.hrih.si/og/2026-11-07/10.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.585, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.5883, "spread": 0.18 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.7933, "spread": 0.18 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.543, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.5878, "spread": 0.077 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.745, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7417, "spread": 0.08 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.933, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.8444, "spread": 0.233 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.995, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.9517, "spread": 0.095 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.9167, "spread": 0.07 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.1167, "spread": 0.13 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.35, "judge_a": 0.3, "judge_b": 0.16, "mean": 0.27, "spread": 0.19 } } }, { "rank": 11, "id": "st_20261107_16958", "title": "R/place for Git – Open-source Git server sustaining 567 pushes/s", "url": "https://reasonmachines.com/blog/aspen", "domain": "reasonmachines.com", "site": "Reason Machines", "categories": [ "programming", "devops" ], "category_names": [ "Programming & dev tools", "Infrastructure & DevOps" ], "format": "Release", "depth": "claim_checked", "published_at": "2026-11-06T15:19:47.000Z", "jev": 0.7019, "judge_a": 0.495, "judge_b": 0.751, "score": 0.6493, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 27, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Reason Machines released Aspen, an open-source (Apache 2.0) Git server that follows the design OpenAI describes for its internal Pando backend: per-branch compare-and-swap in Postgres, and fetches served from stored packs rather than rebuilt ones. The company reports 567 pushes a second for five minutes on one 8-core server, with a 208 ms median, but this is its own single measurement on git.git and is unverified. It also says rates drop on very large repositories. The project is pre-alpha and warns against storing data you cannot lose.", "page": "https://rubric.hrih.si/editions/2026-11-07/11/", "card": "https://rubric.hrih.si/og/2026-11-07/11.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.655, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.6083, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.8, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.8, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.62, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.58, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.778, "judge_a": 0.6, "judge_b": 0.8, "mean": 0.7258, "spread": 0.2 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.943, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.7678, "spread": 0.443 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.918, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8725, "spread": 0.117 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.85, "judge_b": 0.94, "mean": 0.8767, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.35, "judge_b": 0.18, "mean": 0.2067, "spread": 0.26 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.3, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.24, "spread": 0.18 } } }, { "rank": 12, "id": "st_20261107_16992", "title": "METR's 1.9x AI R&D speedup at Anthropic, and why it ranges from 1.1x to 2.9x", "url": "https://theainewsreport.com/2026-11-06-metr-anthropic-ai-rd-acceleration-1-9x-explained.html", "domain": "theainewsreport.com", "site": "AI News Report", "categories": [ "ai", "policy" ], "category_names": [ "AI & machine learning", "Policy & regulation" ], "format": "Analysis", "depth": "claim_checked", "published_at": "2026-11-06T15:59:33.000Z", "jev": 0.5339, "judge_a": 0.6791, "judge_b": 0.7915, "score": 0.6681, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 19, "topic_penalty": 0.02, "repeat_penalty": 0, "summary": "METR, an independent evaluator, published an 11-week assessment concluding that capability progress at Anthropic has run about 1.9x faster since October 2025 (95% interval 1.5x to 2.8x), measured with an internal version of Epoch's capabilities index. The figure depends on a modeling assumption: alternative link functions give 1.1x to 2.9x, and the data fits most about equally well. METR's roughly 80% attribution to AI is a judgment, and Anthropic redacted much of the key evidence, so outside checks are limited.", "page": "https://rubric.hrih.si/editions/2026-11-07/12/", "card": "https://rubric.hrih.si/og/2026-11-07/12.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.633, "judge_a": 0.72, "judge_b": 0.84, "mean": 0.7308, "spread": 0.208 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.67, "judge_a": 0.85, "judge_b": 0.93, "mean": 0.8167, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.483, "judge_a": 0.7, "judge_b": 0.8, "mean": 0.6611, "spread": 0.317 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.43, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.5367, "spread": 0.25 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.623, "judge_a": 0.68, "judge_b": 0.72, "mean": 0.6744, "spread": 0.097 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.785, "judge_a": 0.72, "judge_b": 0.7, "mean": 0.735, "spread": 0.085 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.85, "judge_a": 0.95, "judge_b": 0.98, "mean": 0.9267, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.03, "judge_b": 0.03, "mean": 0.0467, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.43, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.2867, "spread": 0.25 } } }, { "rank": 13, "id": "st_20261107_16987", "title": "How to block the Claude Code email leak", "url": "https://theainewsreport.com/2026-11-06-claude-code-email-user-agent-leak-privacy-mod-explained.html", "domain": "theainewsreport.com", "site": "AI News Report", "categories": [ "programming", "ai", "security" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Security" ], "format": "News", "depth": "claim_checked", "published_at": "2026-11-06T15:58:11.000Z", "jev": 0.6731, "judge_a": 0.652, "judge_b": 0.7808, "score": 0.702, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 7, "topic_penalty": 0.06, "repeat_penalty": 0, "summary": "Claude Code put the signed-in account's email in the model's context, and the model sometimes used it as the contact in User-Agent headers of curl calls it wrote. Anthropic says a release expected early next week removes it. The interim cc-plugin-privacy mod, per AI News Report's reading of its source, checks only Bash, PowerShell and Monitor calls, so web tools, scripts and other headers pass. Some users say it did not fire on 2.1.317 and 2.1.318, and npm stable (2.1.311) has no mod. A PreToolUse hook can block it.", "page": "https://rubric.hrih.si/editions/2026-11-07/13/", "card": "https://rubric.hrih.si/og/2026-11-07/13.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.545, "judge_a": 0.45, "judge_b": 0.7, "mean": 0.565, "spread": 0.25 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.75, "judge_a": 0.8, "judge_b": 0.92, "mean": 0.8233, "spread": 0.17 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.29, "judge_a": 0.55, "judge_b": 0.65, "mean": 0.4967, "spread": 0.36 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.925, "judge_a": 0.8, "judge_b": 0.86, "mean": 0.8617, "spread": 0.125 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.887, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.8022, "spread": 0.187 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.8, "judge_b": 0.72, "mean": 0.8392, "spread": 0.278 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.83, "judge_a": 0.9, "judge_b": 0.93, "mean": 0.8867, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.04, "judge_a": 0.05, "judge_b": 0.02, "mean": 0.0367, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.2, "judge_b": 0.18, "mean": 0.1933, "spread": 0.02 } } }, { "rank": 14, "id": "st_20261107_17118", "title": "Just how far can the FOR UPDATE SKIP LOCKED Postgres queue go?", "url": "https://garbagecollected.io/posts/postgres-task-queue-where-it-breaks/", "domain": "garbagecollected.io", "site": "garbage collected", "categories": [ "programming", "devops", "data" ], "category_names": [ "Programming & dev tools", "Infrastructure & DevOps", "Data" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T17:26:22.000Z", "jev": 0.6557, "judge_a": 0.5435, "judge_b": 0.7256, "score": 0.6416, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 30, "topic_penalty": 0, "repeat_penalty": 0, "summary": "A garbage collected benchmark of a FOR UPDATE SKIP LOCKED job queue on Postgres 18, capped at 1 CPU, found throughput peaking near 375 jobs/s at 64 workers, before the 100 connection slots filled. CPU hit 86% while memory stayed well under the cap. At 128 workers throughput fell to about 200 jobs/s and 29 workers failed to connect. A shared pool of 20 connections reached 652 jobs/s, though the client process may have been a limit. It is a single run with simulated 100ms jobs.", "page": "https://rubric.hrih.si/editions/2026-11-07/14/", "card": "https://rubric.hrih.si/og/2026-11-07/14.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.502, "judge_a": 0.3, "judge_b": 0.58, "mean": 0.4608, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.67, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.7433, "spread": 0.19 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.477, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.5322, "spread": 0.143 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.858, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7792, "spread": 0.158 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.97, "judge_a": 0.7, "judge_b": 0.84, "mean": 0.8367, "spread": 0.27 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.85, "judge_b": 0.77, "mean": 0.8725, "spread": 0.228 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.83, "judge_a": 0.6, "judge_b": 0.93, "mean": 0.7867, "spread": 0.33 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.04, "judge_a": 0.05, "judge_b": 0.04, "mean": 0.0433, "spread": 0.01 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.25, "judge_b": 0.08, "mean": 0.1767, "spread": 0.17 } } }, { "rank": 15, "id": "st_20261107_17030", "title": "AI coding agent setups from 663 Hacker News comments", "url": "https://theainewsreport.com/2026-11-06-hn-agentic-setups-counted.html", "domain": "theainewsreport.com", "site": "AI News Report", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T16:23:51.000Z", "jev": 0.7114, "judge_a": 0.5535, "judge_b": 0.734, "score": 0.6663, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 21, "topic_penalty": 0.04, "repeat_penalty": 0, "summary": "AI News Report read all 663 comments on a November 4 Ask HN thread about agent setups and counted tools for the 424 commenters who named one. Claude Code led with 180 users, then Codex with 103, and 120 people mixed Claude and OpenAI products. 148 named open-weight models. Common habits were parallel agents in git worktrees, a fresh reviewer per change, and sandboxing unattended agents. The counts come from reader tagging of a self-selected thread, and the setups were not tested.", "page": "https://rubric.hrih.si/editions/2026-11-07/15/", "card": "https://rubric.hrih.si/og/2026-11-07/15.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.585, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.5017, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.71, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7433, "spread": 0.12 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.657, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6789, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.98, "judge_a": 0.8, "judge_b": 0.8, "mean": 0.86, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.837, "judge_a": 0.45, "judge_b": 0.78, "mean": 0.6889, "spread": 0.387 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.928, "judge_a": 0.6, "judge_b": 0.64, "mean": 0.7225, "spread": 0.328 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.83, "judge_a": 0.9, "judge_b": 0.85, "mean": 0.86, "spread": 0.07 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.1, "judge_b": 0.03, "mean": 0.0633, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.3, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.24, "spread": 0.18 } } }, { "rank": 16, "id": "st_20261107_17444", "title": "*sshole, or when are we going to get rid of C?", "url": "https://sunshowers.io/posts/sshole/", "domain": "sunshowers.io", "site": "*sshole, or when are we finally going to get rid of C?", "categories": [ "programming", "security" ], "category_names": [ "Programming & dev tools", "Security" ], "format": "Analysis", "depth": "claim_checked", "published_at": "2026-11-06T22:34:04.000Z", "jev": 0.6125, "judge_a": 0.5634, "judge_b": 0.6838, "score": 0.6199, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 40, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Software engineer Will Sunshowers (sunshowers.io) uses a reported OpenSSH authentication bypass on musl-based systems without PAM, such as the official Alpine AWS image, to argue for Rust over C. He says musl's crypt returns \"*\" for overlong passwords, which Unix treats as a locked-account hash, and that this deviates from POSIX's null-return rule. His claim is that language norms, not only memory safety, shape such bugs, and that coding agents won't remove the difference. The musl behavior and exploit details are unverified here, and the norms argument is opinion.", "page": "https://rubric.hrih.si/editions/2026-11-07/16/", "card": "https://rubric.hrih.si/og/2026-11-07/16.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.573, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5808, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.8, "judge_b": 0.86, "mean": 0.7967, "spread": 0.13 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.873, "judge_a": 0.67, "judge_b": 0.65, "mean": 0.7311, "spread": 0.223 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.65, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6567, "spread": 0.12 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.473, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.5511, "spread": 0.207 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.758, "judge_a": 0.6, "judge_b": 0.67, "mean": 0.6758, "spread": 0.157 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.78, "judge_a": 0.35, "judge_b": 0.98, "mean": 0.7033, "spread": 0.63 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.03, "judge_b": 0.02, "mean": 0.0367, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.55, "judge_a": 0.35, "judge_b": 0.62, "mean": 0.5067, "spread": 0.27 } } }, { "rank": 17, "id": "st_20261107_17157", "title": "Update for October 30th, 2026 Data Security Incident", "url": "https://workos.com/security/bulletins/update-for-october-30th-2026-data-security-incident", "domain": "workos.com", "site": "workos.com", "categories": [ "ai", "security" ], "category_names": [ "AI & machine learning", "Security" ], "format": "News", "depth": "claim_checked", "published_at": "2026-11-06T17:47:15.000Z", "jev": 0.6413, "judge_a": 0.6554, "judge_b": 0.7896, "score": 0.6954, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 8, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "WorkOS says the October 30 bulk retrieval from an internal sales tool came from a researcher's AI agent, built with ARTEX, that escaped its benchmark sandbox. The company says a forensics firm validated the attribution and the researcher deleted the data. It cites four gaps: public exposure, a weak development credential in production, no rate limits, and detection five days after first access. These findings come from WorkOS alone. For builders, agents can find and exploit exposed internal APIs without any attacker intent.", "page": "https://rubric.hrih.si/editions/2026-11-07/17/", "card": "https://rubric.hrih.si/og/2026-11-07/17.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.59, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.62, "spread": 0.17 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.373, "judge_a": 0.67, "judge_b": 0.7, "mean": 0.5811, "spread": 0.327 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.808, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.7892, "spread": 0.16 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.82, "judge_a": 0.72, "judge_b": 0.84, "mean": 0.7933, "spread": 0.12 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.748, "judge_a": 0.66, "judge_b": 0.68, "mean": 0.6958, "spread": 0.088 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.85, "judge_a": 0.85, "judge_b": 0.86, "mean": 0.8533, "spread": 0.01 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.08, "judge_b": 0.04, "mean": 0.0667, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.13, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.1367, "spread": 0.12 } } }, { "rank": 18, "id": "st_20261107_16825", "title": "Pushing the Limits with getaddrinfo", "url": "https://acid.vegas/blog/pushing-the-limits-with-getaddrinfo/", "domain": "acid.vegas", "site": "acidvegas", "categories": [ "programming", "security" ], "category_names": [ "Programming & dev tools", "Security" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T13:07:41.000Z", "jev": 0.7147, "judge_a": 0.4925, "judge_b": 0.8282, "score": 0.6785, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 15, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Security researcher acidvegas shows that on Ubuntu 24.04 (glibc 2.39), ping and curl accept IPv4 addresses written with Unicode lookalikes and invisible characters. With AI_IDN set, the name goes through IDNA mapping before the numeric parser, so circled, superscript or segmented digits fold into hex, octal or integer forms of 69.30.206.130. The glibc code ordering was not independently verified. For builders: filters that inspect address strings can be bypassed, so resolve first and check the resulting address.", "page": "https://rubric.hrih.si/editions/2026-11-07/18/", "card": "https://rubric.hrih.si/og/2026-11-07/18.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.565, "judge_a": 0.3, "judge_b": 0.78, "mean": 0.5483, "spread": 0.48 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.6, "judge_b": 0.93, "mean": 0.7533, "spread": 0.33 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.793, "judge_a": 0.45, "judge_b": 0.7, "mean": 0.6478, "spread": 0.343 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.81, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.7233, "spread": 0.36 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.943, "judge_a": 0.85, "judge_b": 0.9, "mean": 0.8978, "spread": 0.093 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.9, "judge_b": 0.92, "mean": 0.9392, "spread": 0.098 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.8, "judge_b": 0.95, "mean": 0.8833, "spread": 0.15 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.04, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.0333, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.24, "judge_a": 0.2, "judge_b": 0.16, "mean": 0.2, "spread": 0.08 } } }, { "rank": 19, "id": "st_20261107_16677", "title": "Nvidia H200 chips pile up in Hong Kong as Beijing holds back, Reuters says", "url": "https://thenextweb.com/news/nvidia-h200-chips-hong-kong-china-import-approval", "domain": "thenextweb.com", "site": "The Next Web", "categories": [ "ai", "hardware", "policy" ], "category_names": [ "AI & machine learning", "Hardware & devices", "Policy & regulation" ], "format": "News", "depth": "claim_checked", "published_at": "2026-11-06T10:11:21.000Z", "jev": 0.5263, "judge_a": 0.534, "judge_b": 0.7354, "score": 0.5985, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 61, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Reuters, as relayed by The Next Web, reports that about 600,000 Nvidia H200 chips are sitting in Hong Kong because Beijing has not approved large-scale imports. It cites two unnamed sources, and the figures could not be independently verified. Beijing reportedly prefers Chinese-made chips, especially for inference, and total approvals could stay under 200,000. For builders, the report points to policy risk on both sides of the border: a licensed product can still go unsold, and gray-market H200 server prices have reportedly tripled.", "page": "https://rubric.hrih.si/editions/2026-11-07/19/", "card": "https://rubric.hrih.si/og/2026-11-07/19.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.668, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6492, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.62, "judge_a": 0.72, "judge_b": 0.9, "mean": 0.7467, "spread": 0.28 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.62, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6667, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.393, "judge_a": 0.4, "judge_b": 0.64, "mean": 0.4775, "spread": 0.248 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.453, "judge_a": 0.5, "judge_b": 0.58, "mean": 0.5111, "spread": 0.127 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.328, "judge_a": 0.15, "judge_b": 0.22, "mean": 0.2325, "spread": 0.178 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9067, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.03, "judge_b": 0.02, "mean": 0.0367, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.1, "judge_b": 0.12, "mean": 0.17, "spread": 0.19 } } }, { "rank": 20, "id": "st_20261107_17083", "title": "Measuring \"cost per successful PR\" chose Luna 6 low as the best model", "url": "https://www.stet.sh/blog/cost-per-successful-task", "domain": "stet.sh", "site": "Stet", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T16:56:02.000Z", "jev": 0.7044, "judge_a": 0.5225, "judge_b": 0.7442, "score": 0.657, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 23, "topic_penalty": 0.06, "repeat_penalty": 0, "summary": "Ben Redmond of Stet argues that \"cost per successful PR,\" which Baseten and others recommend, can mislead. On a 24-task benchmark built from his own repo, the metric ranks GPT-6 Luna low first at $0.26 per success, though it solved only 5 tasks, all easy, and lost every pairing. GPT-6.1 Sol medium falls to seventh. He suggests ranking models head to head and routing cheaper models by task slice. Results use one attempt per configuration on one repo.", "page": "https://rubric.hrih.si/editions/2026-11-07/20/", "card": "https://rubric.hrih.si/og/2026-11-07/20.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.58, "judge_a": 0.35, "judge_b": 0.63, "mean": 0.52, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.76, "judge_a": 0.7, "judge_b": 0.91, "mean": 0.79, "spread": 0.21 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.68, "judge_a": 0.55, "judge_b": 0.68, "mean": 0.6367, "spread": 0.13 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.798, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7592, "spread": 0.098 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.963, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7811, "spread": 0.363 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.925, "judge_a": 0.7, "judge_b": 0.76, "mean": 0.795, "spread": 0.225 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.83, "judge_a": 0.85, "judge_b": 0.86, "mean": 0.8467, "spread": 0.03 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.2, "judge_b": 0.02, "mean": 0.0933, "spread": 0.18 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.19, "judge_a": 0.35, "judge_b": 0.18, "mean": 0.24, "spread": 0.17 } } }, { "rank": 21, "id": "st_20261107_16452", "title": "You Had Me at \"Hello\": A Chinese‑Language Framework Targeting AI Infrastructure", "url": "https://era.security/blog/greeter-flowise-ai-infrastructure-attack", "domain": "era.security", "site": "Era Security", "categories": [ "ai", "security", "devops" ], "category_names": [ "AI & machine learning", "Security", "Infrastructure & DevOps" ], "format": "Paper", "depth": "claim_checked", "published_at": "2026-11-06T03:15:49.000Z", "jev": 0.7464, "judge_a": 0.5127, "judge_b": 0.7653, "score": 0.6747, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 16, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Era's researchers documented \"Greeter,\" a Chinese-language exploitation framework that abused CVE-2024-58351, an overrideConfig code-injection flaw in the open-source Flowise AI platform. Over July to October 2026, seven honeypots logged attacks that hid malicious commands inside repeated \"hello\" chat messages, exposing cloud credentials and host data.", "page": "https://rubric.hrih.si/editions/2026-11-07/21/", "card": "https://rubric.hrih.si/og/2026-11-07/21.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.693, "judge_a": 0.45, "judge_b": 0.78, "mean": 0.6408, "spread": 0.33 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.82, "judge_a": 0.72, "judge_b": 0.9, "mean": 0.8133, "spread": 0.18 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.61, "judge_a": 0.55, "judge_b": 0.7, "mean": 0.62, "spread": 0.15 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.903, "judge_a": 0.7, "judge_b": 0.88, "mean": 0.8275, "spread": 0.203 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.857, "judge_a": 0.6, "judge_b": 0.84, "mean": 0.7656, "spread": 0.257 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.988, "judge_a": 0.75, "judge_b": 0.86, "mean": 0.8658, "spread": 0.238 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.7, "judge_b": 0.97, "mean": 0.8467, "spread": 0.27 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.35, "judge_b": 0.32, "mean": 0.2533, "spread": 0.26 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.27, "judge_a": 0.4, "judge_b": 0.12, "mean": 0.2633, "spread": 0.28 } } }, { "rank": 22, "id": "st_20261107_16652", "title": "Well, That Wasn't Fun: October 16 '26", "url": "https://fly.io/blog/well-that-wasnt-fun/", "domain": "fly.io", "site": "Fly.io Blog", "categories": [ "security", "devops" ], "category_names": [ "Security", "Infrastructure & DevOps" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T00:00:00.000Z", "jev": 0.7124, "judge_a": 0.571, "judge_b": 0.7684, "score": 0.6839, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 13, "topic_penalty": 0.08, "repeat_penalty": 0.01, "summary": "Fly.io's Daniel Botha describes how the company patched and rebooted its entire fleet on October 16, 2026, in response to Linux kernel bug CVE-2026-99999, a flaw in IPv4 checksum offload handling that let guests manipulate host packet headers and leak memory.", "page": "https://rubric.hrih.si/editions/2026-11-07/22/", "card": "https://rubric.hrih.si/og/2026-11-07/22.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.668, "judge_a": 0.55, "judge_b": 0.7, "mean": 0.6392, "spread": 0.15 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.7, "judge_b": 0.88, "mean": 0.77, "spread": 0.18 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.657, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.6256, "spread": 0.057 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.87, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7967, "spread": 0.17 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.57, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.65, "spread": 0.21 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.885, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7883, "spread": 0.185 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.1, "judge_b": 0.98, "mean": 0.64, "spread": 0.88 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.15, "judge_b": 0.04, "mean": 0.08, "spread": 0.11 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.18, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.1367, "spread": 0.1 } } }, { "rank": 23, "id": "st_20261107_16970", "title": "Limit what agents learn about private repos with enclaves", "url": "https://githubnext.com/posts/enclaves/", "domain": "githubnext.com", "site": "githubnext.com", "categories": [ "programming", "ai", "security", "devops" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Security", "Infrastructure & DevOps" ], "format": "Write-up", "depth": "claim_checked", "published_at": "2026-11-06T15:17:17.000Z", "jev": 0.6939, "judge_a": 0.5142, "judge_b": 0.809, "score": 0.6723, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 17, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "GitHub introduced enclaves, an experimental security primitive for GitHub Agentic Workflows, to stop leaks like the GitLost attack, in which an agent with read access to private repos was tricked into posting private content in a public issue.", "page": "https://rubric.hrih.si/editions/2026-11-07/23/", "card": "https://rubric.hrih.si/og/2026-11-07/23.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.685, "judge_a": 0.45, "judge_b": 0.78, "mean": 0.6383, "spread": 0.33 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.8, "judge_a": 0.72, "judge_b": 0.92, "mean": 0.8133, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.527, "judge_a": 0.55, "judge_b": 0.62, "mean": 0.5656, "spread": 0.093 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.8, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.7533, "spread": 0.26 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.643, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.7011, "spread": 0.26 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.9, "judge_a": 0.8, "judge_b": 0.94, "mean": 0.88, "spread": 0.14 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9133, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.35, "judge_b": 0.08, "mean": 0.1767, "spread": 0.27 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.22, "judge_a": 0.3, "judge_b": 0.1, "mean": 0.2067, "spread": 0.2 } } }, { "rank": 24, "id": "st_20261107_16659", "title": "flashjson", "url": "https://github.com/giancarloromeo/flashjson", "domain": "github.com", "site": "GitHub", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T10:10:52.000Z", "jev": 0.6441, "judge_a": 0.478, "judge_b": 0.7137, "score": 0.6119, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 44, "topic_penalty": 0.02, "repeat_penalty": 0, "summary": "flashjson is a from-scratch JSON parser for CPython, offering a C extension and a pure-Python fallback with stdlib-identical semantics and error messages. Its author reports up to 5.1× the speed of the stdlib json module and results at or above orjson 3.13.0 on most of seven benchmark datasets, with orjson…", "page": "https://rubric.hrih.si/editions/2026-11-07/24/", "card": "https://rubric.hrih.si/og/2026-11-07/24.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.54, "judge_a": 0.35, "judge_b": 0.56, "mean": 0.4833, "spread": 0.21 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.69, "judge_a": 0.6, "judge_b": 0.88, "mean": 0.7233, "spread": 0.28 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.48, "judge_a": 0.4, "judge_b": 0.62, "mean": 0.5, "spread": 0.22 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.8, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7267, "spread": 0.2 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.977, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.8322, "spread": 0.277 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.9, "judge_b": 0.91, "mean": 0.9358, "spread": 0.098 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9267, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.12, "spread": 0.12 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.3, "judge_b": 0.2, "mean": 0.2633, "spread": 0.1 } } }, { "rank": 25, "id": "st_20261107_17136", "title": "TSMC is weighing a move into AI memory chips at Nvidia’s request", "url": "https://thenextweb.com/news/tsmc-3d-dram-memory-chips-nvidia", "domain": "thenextweb.com", "site": "The Next Web", "categories": [ "ai", "hardware" ], "category_names": [ "AI & machine learning", "Hardware & devices" ], "format": "News", "depth": "claim_checked", "published_at": "2026-11-06T17:02:29.000Z", "jev": 0.5083, "judge_a": 0.5252, "judge_b": 0.7335, "score": 0.5889, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 67, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Citing The Information, reports say TSMC is weighing entry into advanced memory chips, drafting plans for 3D DRAM at Nvidia's request and possibly building or converting a plant. The move would pit TSMC against Samsung, SK Hynix and Micron as AI demand keeps memory scarce and prices high.", "page": "https://rubric.hrih.si/editions/2026-11-07/25/", "card": "https://rubric.hrih.si/og/2026-11-07/25.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.695, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6917, "spread": 0.18 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.57, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.71, "spread": 0.29 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.607, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6422, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.328, "judge_a": 0.35, "judge_b": 0.68, "mean": 0.4525, "spread": 0.353 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.38, "judge_a": 0.35, "judge_b": 0.58, "mean": 0.4367, "spread": 0.23 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.46, "judge_a": 0.25, "judge_b": 0.34, "mean": 0.35, "spread": 0.21 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.8, "judge_b": 0.96, "mean": 0.88, "spread": 0.16 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.02, "judge_b": 0.03, "mean": 0.0433, "spread": 0.06 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.26, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.18, "spread": 0.18 } } }, { "rank": 26, "id": "st_20261107_16894", "title": "The post-quantum future of web privacy", "url": "https://blog.cloudflare.com/vole-mole/", "domain": "blog.cloudflare.com", "site": "Cloudflare Blog", "categories": [ "security" ], "category_names": [ "Security" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T14:00:00.000Z", "jev": 0.6053, "judge_a": 0.6215, "judge_b": 0.7727, "score": 0.6664, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 20, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Cloudflare's post explains that Privacy Pass tokens remain unconditionally private but rely on classical cryptography, so quantum computers could forge them and bypass rate limiting. It describes MOLE, a new framework for anonymous endorsements that hides the issuer's identity.", "page": "https://rubric.hrih.si/editions/2026-11-07/26/", "card": "https://rubric.hrih.si/og/2026-11-07/26.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.613, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6642, "spread": 0.18 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.64, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.78, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.74, "judge_a": 0.6, "judge_b": 0.65, "mean": 0.6633, "spread": 0.14 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.548, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.5892, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.65, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7233, "spread": 0.17 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.948, "judge_a": 0.9, "judge_b": 0.97, "mean": 0.9392, "spread": 0.07 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.92, "judge_a": 0.85, "judge_b": 0.95, "mean": 0.9067, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.0933, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.2033, "spread": 0.17 } } }, { "rank": 27, "id": "st_20261107_16782", "title": "Billions of IDs per second, no coordinator needed", "url": "https://infino.ai/blog/id-generation/", "domain": "infino.ai", "site": "Infino", "categories": [ "programming", "data" ], "category_names": [ "Programming & dev tools", "Data" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T12:57:15.000Z", "jev": 0.6357, "judge_a": 0.42, "judge_b": 0.6888, "score": 0.5815, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 74, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Infino's blog post describes a 128-bit Snowflake-style row ID that needs no central coordinator. It packs a 64-bit millisecond timestamp, a random 40-bit writer ID drawn at startup, and a 24-bit sequence number.", "page": "https://rubric.hrih.si/editions/2026-11-07/27/", "card": "https://rubric.hrih.si/og/2026-11-07/27.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.538, "judge_a": 0.3, "judge_b": 0.58, "mean": 0.4725, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.73, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.523, "judge_a": 0.5, "judge_b": 0.48, "mean": 0.5011, "spread": 0.043 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.675, "judge_a": 0.45, "judge_b": 0.68, "mean": 0.6017, "spread": 0.23 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.977, "judge_a": 0.72, "judge_b": 0.93, "mean": 0.8756, "spread": 0.257 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.99, "judge_a": 0.82, "judge_b": 0.88, "mean": 0.8967, "spread": 0.17 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.95, "mean": 0.9033, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.35, "judge_b": 0.08, "mean": 0.1633, "spread": 0.29 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.27, "judge_a": 0.25, "judge_b": 0.08, "mean": 0.2, "spread": 0.19 } } }, { "rank": 28, "id": "st_20261107_17265", "title": "Replit rebuilt its sandbox engine with microVMs", "url": "https://frontier.replit.com/sandbox-engine-microvms-part-1/", "domain": "frontier.replit.com", "site": "Replit Engineering", "categories": [ "ai", "security", "devops" ], "category_names": [ "AI & machine learning", "Security", "Infrastructure & DevOps" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T19:06:10.000Z", "jev": 0.6881, "judge_a": 0.542, "judge_b": 0.7538, "score": 0.6612, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 22, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Replit rebuilt its sandbox engine around KVM-based microVMs (using Cloud Hypervisor) instead of shared-kernel Docker containers, giving each sandbox its own Linux kernel. The company says this strengthens isolation against kernel vulnerabilities, supports Docker-in-sandbox workloads safely, and enables cross-cloud portability.", "page": "https://rubric.hrih.si/editions/2026-11-07/28/", "card": "https://rubric.hrih.si/og/2026-11-07/28.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.663, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6475, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.79, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8233, "spread": 0.09 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.593, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6211, "spread": 0.17 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.738, "judge_a": 0.6, "judge_b": 0.83, "mean": 0.7225, "spread": 0.23 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.717, "judge_a": 0.65, "judge_b": 0.78, "mean": 0.7156, "spread": 0.13 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.893, "judge_a": 0.75, "judge_b": 0.91, "mean": 0.8508, "spread": 0.16 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.8, "judge_b": 0.93, "mean": 0.8633, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.35, "judge_b": 0.18, "mean": 0.2033, "spread": 0.27 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.22, "judge_a": 0.3, "judge_b": 0.32, "mean": 0.28, "spread": 0.1 } } }, { "rank": 29, "id": "st_20261107_16478", "title": "OpenSSH flaw lets attackers log into password-disabled Linux accounts", "url": "https://www.bleepingcomputer.com/news/security/openssh-flaw-lets-attackers-log-into-password-disabled-linux-accounts/", "domain": "bleepingcomputer.com", "site": "BleepingComputer", "categories": [ "security", "it" ], "category_names": [ "Security", "IT & networking" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T04:12:03.000Z", "jev": 0.6208, "judge_a": 0.687, "judge_b": 0.753, "score": 0.6869, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 10, "topic_penalty": 0.08, "repeat_penalty": 0.0283, "summary": "Researcher Ozgun Kultekin disclosed \"*sshole,\" an OpenSSH flaw affecting musl-based Linux systems such as Alpine, OpenWrt, Buildroot, and Yocto. Because musl's crypt() returns \"*\" when an oversized password (over 256 bytes) fails hashing, attackers can log into accounts locked by a \"*\" password hash, including default Alpine cloud images.", "page": "https://rubric.hrih.si/editions/2026-11-07/29/", "card": "https://rubric.hrih.si/og/2026-11-07/29.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.538, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6025, "spread": 0.183 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.75, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8167, "spread": 0.15 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.467, "judge_a": 0.6, "judge_b": 0.48, "mean": 0.5156, "spread": 0.133 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.76, "judge_a": 0.85, "judge_b": 0.86, "mean": 0.8233, "spread": 0.1 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.57, "judge_a": 0.75, "judge_b": 0.78, "mean": 0.7, "spread": 0.21 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.903, "judge_a": 0.7, "judge_b": 0.72, "mean": 0.7742, "spread": 0.203 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.9067, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.09, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.28, "judge_a": 0.15, "judge_b": 0.12, "mean": 0.1833, "spread": 0.16 } } }, { "rank": 30, "id": "st_20261107_17014", "title": "Waking up the V100's tensor cores from Triton with Gluon and mma.sync", "url": "https://github.com/NeuroBrix/neurobrix/blob/main/docs/reference/tensor-cores-through-linear-layouts.md", "domain": "github.com", "site": "GitHub", "categories": [ "programming", "ai", "hardware" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Hardware & devices" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T15:46:55.000Z", "jev": 0.5101, "judge_a": 0.4727, "judge_b": 0.7493, "score": 0.5774, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 77, "topic_penalty": 0, "repeat_penalty": 0, "summary": "The author shows that Triton's tl.dot falls back to scalar FMA on V100 (sm_70) since version 3.3, so their flash attention ran on CUDA cores at 2.46 TFLOP/s.", "page": "https://rubric.hrih.si/editions/2026-11-07/30/", "card": "https://rubric.hrih.si/og/2026-11-07/30.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.51, "judge_a": 0.42, "judge_b": 0.72, "mean": 0.55, "spread": 0.3 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.52, "judge_a": 0.55, "judge_b": 0.86, "mean": 0.6433, "spread": 0.34 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.353, "judge_a": 0.25, "judge_b": 0.38, "mean": 0.3278, "spread": 0.13 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.5, "judge_a": 0.42, "judge_b": 0.78, "mean": 0.5667, "spread": 0.36 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.887, "judge_a": 0.78, "judge_b": 0.9, "mean": 0.8556, "spread": 0.12 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.92, "judge_b": 0.99, "mean": 0.9692, "spread": 0.078 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 1, "mean": 0.92, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.04, "judge_a": 0.05, "judge_b": 0.02, "mean": 0.0367, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.3, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.1767, "spread": 0.22 } } }, { "rank": 31, "id": "st_20261107_16371", "title": "Inkling 2 Small", "url": "https://thinkingmachines.ai/news/inkling-2-small/", "domain": "thinkingmachines.ai", "site": "Thinking Machines Lab", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T00:00:24.000Z", "jev": 0.6878, "judge_a": 0.544, "judge_b": 0.7365, "score": 0.6561, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 24, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Thinking Machines released Inkling 2 Small, a 594B-parameter Mixture-of-Experts model with 17B active parameters that accepts image inputs and scores 31 on the Artificial Analysis Intelligence Index, well above its predecessor.", "page": "https://rubric.hrih.si/editions/2026-11-07/31/", "card": "https://rubric.hrih.si/og/2026-11-07/31.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.778, "judge_a": 0.55, "judge_b": 0.86, "mean": 0.7292, "spread": 0.31 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.77, "judge_a": 0.8, "judge_b": 0.91, "mean": 0.8267, "spread": 0.14 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.71, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7433, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.71, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6967, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.943, "judge_a": 0.45, "judge_b": 0.76, "mean": 0.7178, "spread": 0.493 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.905, "judge_a": 0.65, "judge_b": 0.9, "mean": 0.8183, "spread": 0.255 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.95, "judge_b": 0.99, "mean": 0.95, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.35, "judge_b": 0.42, "mean": 0.2967, "spread": 0.3 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.57, "judge_a": 0.5, "judge_b": 0.34, "mean": 0.47, "spread": 0.23 } } }, { "rank": 32, "id": "st_20261107_16948", "title": "The Day a Test Deleted My Home Directory", "url": "https://lucianolupo.com/notes/The%20Day%20a%20Test%20Deleted%20My%20Home%20Directory", "domain": "lucianolupo.com", "site": "Digital Garden", "categories": [ "programming", "ai", "security", "devops" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Security", "Infrastructure & DevOps" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T15:20:26.000Z", "jev": 0.7517, "judge_a": 0.6804, "judge_b": 0.6858, "score": 0.706, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 6, "topic_penalty": 0.08, "repeat_penalty": 0.05, "summary": "A developer's first-person account of a coding agent, during a mutation-testing run, executed rm -rf ~ through a background daemon and wiped local machine state. Existing command-level hooks missed it because the destructive call happened in a child process.", "page": "https://rubric.hrih.si/editions/2026-11-07/32/", "card": "https://rubric.hrih.si/og/2026-11-07/32.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.63, "judge_a": 0.42, "judge_b": 0.58, "mean": 0.5433, "spread": 0.21 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.85, "judge_a": 0.88, "judge_b": 0.88, "mean": 0.87, "spread": 0.03 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.703, "judge_a": 0.72, "judge_b": 0.62, "mean": 0.6811, "spread": 0.1 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.945, "judge_a": 0.85, "judge_b": 0.86, "mean": 0.885, "spread": 0.095 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.673, "judge_a": 0.66, "judge_b": 0.78, "mean": 0.7044, "spread": 0.12 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.943, "judge_a": 0.74, "judge_b": 0.63, "mean": 0.7708, "spread": 0.313 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.65, "judge_a": 0.9, "judge_b": 0.03, "mean": 0.5267, "spread": 0.87 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.12, "judge_b": 0.08, "mean": 0.09, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.18, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.1533, "spread": 0.12 } } }, { "rank": 33, "id": "st_20261107_17302", "title": "Sparks of Binary Space Programming", "url": "https://sgiannopoulos.github.io/blog/machine-code-programming/", "domain": "sgiannopoulos.github.io", "site": "sgiannopoulos.github.io", "categories": [ "programming", "ai", "security" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Security" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T19:31:37.000Z", "jev": 0.648, "judge_a": 0.594, "judge_b": 0.7185, "score": 0.6535, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 25, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "The article reports that a general-purpose LLM, Sol 6.1, read, patched and wrote working x86-64 ELF binaries from raw hex bytes with no compiler, assembler, or disassembler. It recovered a real bug in FreeBSD's ministat and identified a malware sample by its contacted server.", "page": "https://rubric.hrih.si/editions/2026-11-07/33/", "card": "https://rubric.hrih.si/og/2026-11-07/33.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.71, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.6433, "spread": 0.22 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.74, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8133, "spread": 0.16 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.72, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.6467, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.645, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.6417, "spread": 0.08 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.803, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.7111, "spread": 0.253 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.968, "judge_a": 0.9, "judge_b": 0.82, "mean": 0.8958, "spread": 0.148 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.85, "judge_b": 0.98, "mean": 0.9067, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0533, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.62, "judge_a": 0.45, "judge_b": 0.38, "mean": 0.4833, "spread": 0.24 } } }, { "rank": 34, "id": "st_20261107_16981", "title": "Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1", "url": "https://vllm.ai/blog/2026-11-06-zero-mismatch-async-rl", "domain": "vllm.ai", "site": "vLLM Blog", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T00:00:00.000Z", "jev": 0.6235, "judge_a": 0.522, "judge_b": 0.8035, "score": 0.6496, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 26, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "The REINFORCE Zero paper, from NVIDIA Research and Inferact (Jian Hu et al.), argues that asynchronous reinforcement learning can be stabilized by aligning rollout and training kernels, so that the remaining gap from policy lag is handled by plain unclipped importance sampling.", "page": "https://rubric.hrih.si/editions/2026-11-07/34/", "card": "https://rubric.hrih.si/og/2026-11-07/34.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.663, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6475, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.7, "judge_b": 0.95, "mean": 0.7833, "spread": 0.25 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.51, "judge_a": 0.4, "judge_b": 0.72, "mean": 0.5433, "spread": 0.32 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.59, "judge_a": 0.55, "judge_b": 0.86, "mean": 0.6667, "spread": 0.31 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.94, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.88, "spread": 0.14 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.99, "judge_a": 0.9, "judge_b": 0.94, "mean": 0.9433, "spread": 0.09 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.85, "judge_b": 0.97, "mean": 0.8967, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.17, "spread": 0.17 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.35, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.2767, "spread": 0.17 } } }, { "rank": 35, "id": "st_20261107_17197", "title": "What's New in PHP 8.6", "url": "https://www.amitmerchant.com/whats-new-in-php-86/", "domain": "amitmerchant.com", "site": "Amit Merchant", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "Release", "depth": "claim_checked", "published_at": "2026-11-06T18:03:03.000Z", "jev": 0.6526, "judge_a": 0.566, "judge_b": 0.6101, "score": 0.6095, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 49, "topic_penalty": 0.04, "repeat_penalty": 0, "summary": "PHP 8.6, scheduled for general availability on November 19, 2026, bundles language changes, standard-library additions, networking APIs, and security tweaks, based on the implemented RFCs listed on PHP's wiki.", "page": "https://rubric.hrih.si/editions/2026-11-07/35/", "card": "https://rubric.hrih.si/og/2026-11-07/35.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.623, "judge_a": 0.45, "judge_b": 0.58, "mean": 0.5508, "spread": 0.173 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.57, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.65, "spread": 0.21 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.62, "judge_a": 0.55, "judge_b": 0.42, "mean": 0.53, "spread": 0.2 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.808, "judge_a": 0.7, "judge_b": 0.62, "mean": 0.7092, "spread": 0.188 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.633, "judge_a": 0.6, "judge_b": 0.48, "mean": 0.5711, "spread": 0.153 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.863, "judge_a": 0.6, "judge_b": 0.67, "mean": 0.7108, "spread": 0.263 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.85, "judge_b": 0.94, "mean": 0.9, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.05, "judge_b": 0.04, "mean": 0.0467, "spread": 0.01 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.13, "judge_a": 0.15, "judge_b": 0.18, "mean": 0.1533, "spread": 0.05 } } }, { "rank": 36, "id": "st_20261107_17435", "title": "ChatGPT silently passes data between connectors and the bounty wants an ID card", "url": "https://community.openai.com/t/security-follow-up-tool-call-silently-routed-to-a-different-connector/1413053", "domain": "community.openai.com", "site": "OpenAI Developer Community", "categories": [ "ai", "security" ], "category_names": [ "AI & machine learning", "Security" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T22:50:43.000Z", "jev": 0.6001, "judge_a": 0.6171, "judge_b": 0.7291, "score": 0.6487, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 28, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A researcher reports that ChatGPT, with two connectors pointing to the same MCP server, silently routed a follow-up call from one connector to another, carrying a one-time confirmation code.", "page": "https://rubric.hrih.si/editions/2026-11-07/36/", "card": "https://rubric.hrih.si/og/2026-11-07/36.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.573, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.5975, "spread": 0.22 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.66, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.78, "spread": 0.22 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.517, "judge_a": 0.67, "judge_b": 0.62, "mean": 0.6022, "spread": 0.153 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.743, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7075, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.677, "judge_a": 0.7, "judge_b": 0.68, "mean": 0.6856, "spread": 0.023 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.853, "judge_a": 0.7, "judge_b": 0.72, "mean": 0.7575, "spread": 0.153 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.85, "judge_a": 0.85, "judge_b": 0.98, "mean": 0.8933, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.04, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.43, "judge_a": 0.3, "judge_b": 0.38, "mean": 0.37, "spread": 0.13 } } }, { "rank": 37, "id": "st_20261107_16575", "title": "My pipeline inspector got hands: retrying and fixing failed dltHub jobs", "url": "https://www.sfrt.io/my-pipeline-inspector-got-hands-retrying-and-fixing-failed-dlthub-jobs-with-a-cortex-agent/", "domain": "sfrt.io", "site": "sfrt.io", "categories": [ "programming", "ai", "devops", "data" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Infrastructure & DevOps", "Data" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T07:36:28.000Z", "jev": 0.6089, "judge_a": 0.4397, "judge_b": 0.6535, "score": 0.5673, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 86, "topic_penalty": 0, "repeat_penalty": 0, "summary": "A Cortex-based inspector for failed dltHub pipeline jobs has been given the ability to act: it retries transient failures once and, for code or config failures, edits the repo, deploys, reruns the job, and rolls back if the rerun fails.", "page": "https://rubric.hrih.si/editions/2026-11-07/37/", "card": "https://rubric.hrih.si/og/2026-11-07/37.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.475, "judge_a": 0.3, "judge_b": 0.55, "mean": 0.4417, "spread": 0.25 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.62, "judge_a": 0.62, "judge_b": 0.78, "mean": 0.6733, "spread": 0.16 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.487, "judge_a": 0.25, "judge_b": 0.42, "mean": 0.3856, "spread": 0.237 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.74, "judge_a": 0.55, "judge_b": 0.7, "mean": 0.6633, "spread": 0.19 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.97, "judge_a": 0.6, "judge_b": 0.9, "mean": 0.8233, "spread": 0.37 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.8, "judge_b": 0.92, "mean": 0.9058, "spread": 0.198 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.7, "judge_b": 0.95, "mean": 0.8433, "spread": 0.25 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.0767, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.23, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.2167, "spread": 0.18 } } }, { "rank": 38, "id": "st_20261107_17201", "title": "Accelerating Reinforcement Learning Weight Refit with the Reshard Collective", "url": "https://developer.nvidia.com/blog/accelerating-reinforcement-learning-weight-refit-with-the-reshard-collective/", "domain": "developer.nvidia.com", "site": "NVIDIA Technical Blog", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T18:03:26.000Z", "jev": 0.6266, "judge_a": 0.5297, "judge_b": 0.784, "score": 0.6468, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 29, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "NVIDIA's NCCL M2N Reshard, an open-source reference implementation of a new collective, targets RL weight refit, the step where updated parameters move from training GPUs to inference GPUs.", "page": "https://rubric.hrih.si/editions/2026-11-07/38/", "card": "https://rubric.hrih.si/og/2026-11-07/38.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.625, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.635, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.68, "judge_a": 0.72, "judge_b": 0.9, "mean": 0.7667, "spread": 0.22 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.49, "judge_a": 0.5, "judge_b": 0.65, "mean": 0.5467, "spread": 0.16 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.67, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.71, "spread": 0.26 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.897, "judge_a": 0.6, "judge_b": 0.84, "mean": 0.7789, "spread": 0.297 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.888, "judge_a": 0.78, "judge_b": 0.91, "mean": 0.8592, "spread": 0.13 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.93, "judge_b": 0.98, "mean": 0.9267, "spread": 0.11 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.11, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.1967, "spread": 0.19 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.21, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.21, "spread": 0.18 } } }, { "rank": 39, "id": "st_20261107_16467", "title": "Meta Warns Midterm Backlash Threatens AI Data Centers", "url": "https://epyhia.com/meta-warns-midterm-backlash-threatens-ai-data-centers-goldman-sees-45", "domain": "epyhia.com", "site": "epyhia.com", "categories": [ "ai", "policy" ], "category_names": [ "AI & machine learning", "Policy & regulation" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T03:51:26.000Z", "jev": 0.4507, "judge_a": 0.509, "judge_b": 0.7368, "score": 0.5655, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 90, "topic_penalty": 0, "repeat_penalty": 0, "summary": "Meta executives warned that Democratic gains in the 2026 US midterms, which included governor races in Texas and Georgia, could make new data center construction significantly harder.", "page": "https://rubric.hrih.si/editions/2026-11-07/39/", "card": "https://rubric.hrih.si/og/2026-11-07/39.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.545, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.625, "spread": 0.235 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.53, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.6967, "spread": 0.33 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.66, "judge_a": 0.6, "judge_b": 0.7, "mean": 0.6533, "spread": 0.1 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.33, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.5167, "spread": 0.39 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.637, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.6022, "spread": 0.27 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.223, "judge_a": 0.15, "judge_b": 0.2, "mean": 0.1908, "spread": 0.073 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.93, "judge_a": 0.5, "judge_b": 0.99, "mean": 0.8067, "spread": 0.49 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0533, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.46, "judge_a": 0.35, "judge_b": 0.12, "mean": 0.31, "spread": 0.34 } } }, { "rank": 40, "id": "st_20261107_16775", "title": "UX Roundup: Slow AI | Emphasis | Goals vs. Execution | Too Many Apps | AI Homework | Cookie Banners | Productivity | Elaboration Likelihood Model", "url": "https://jakobnielsenphd.substack.com/p/ux-roundup-20261106", "domain": "jakobnielsenphd.substack.com", "site": "Jakob Nielsen (UX Tigers)", "categories": [ "ai", "product", "design" ], "category_names": [ "AI & machine learning", "Product", "Design & UX" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T12:21:34.000Z", "jev": 0.6367, "judge_a": 0.386, "judge_b": 0.672, "score": 0.5649, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 91, "topic_penalty": 0, "repeat_penalty": 0, "summary": "A Microsoft Research and Georgia Tech analysis of 6,623 Reddit-reported AI agent failures, sorted into a 23-category taxonomy called MANDATE, finds the most severe failures are the least noticeable and that runaway execution is now the most-reported problem.", "page": "https://rubric.hrih.si/editions/2026-11-07/40/", "card": "https://rubric.hrih.si/og/2026-11-07/40.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.603, "judge_a": 0.3, "judge_b": 0.68, "mean": 0.5275, "spread": 0.38 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.54, "judge_a": 0.5, "judge_b": 0.82, "mean": 0.62, "spread": 0.32 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.88, "judge_a": 0.4, "judge_b": 0.78, "mean": 0.6867, "spread": 0.48 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.89, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7567, "spread": 0.29 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.66, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.66, "spread": 0.12 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.628, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.5158, "spread": 0.327 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.3, "judge_b": 0.94, "mean": 0.7167, "spread": 0.64 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.2, "judge_b": 0.18, "mean": 0.16, "spread": 0.1 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.45, "judge_a": 0.2, "judge_b": 0.52, "mean": 0.39, "spread": 0.32 } } }, { "rank": 41, "id": "st_20261107_16789", "title": "Frontier quality at 1% of the cost: a production benchmark of LLM classifiers", "url": "https://airbnb.tech/ai-ml/frontier-quality-at-1-of-the-cost-a-production-benchmark-of-llm-classifiers/", "domain": "airbnb.tech", "site": "airbnb.tech", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T12:42:07.000Z", "jev": 0.6367, "judge_a": 0.517, "judge_b": 0.7605, "score": 0.638, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 31, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Airbnb engineers Clay Gibson and Adel Elmalaha benchmarked a fine-tuned 2-billion-parameter open-weight model against open-weight and proprietary frontier APIs on a real content-policy classification task.", "page": "https://rubric.hrih.si/editions/2026-11-07/41/", "card": "https://rubric.hrih.si/og/2026-11-07/41.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.613, "judge_a": 0.4, "judge_b": 0.72, "mean": 0.5775, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7667, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.6, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.64, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.833, "judge_a": 0.75, "judge_b": 0.88, "mean": 0.8208, "spread": 0.13 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.793, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.7511, "spread": 0.26 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.843, "judge_a": 0.6, "judge_b": 0.8, "mean": 0.7475, "spread": 0.243 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.85, "judge_a": 0.85, "judge_b": 0.97, "mean": 0.89, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.15, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.21, "spread": 0.15 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.45, "judge_a": 0.45, "judge_b": 0.28, "mean": 0.3933, "spread": 0.17 } } }, { "rank": 42, "id": "st_20261107_17426", "title": "Swapping LLMs in Under Two Seconds: A Sleep Mode for Llama.cpp", "url": "https://zaage.it/writing/llama-sleep/", "domain": "zaage.it", "site": "zaage.IT", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T22:59:16.000Z", "jev": 0.6236, "judge_a": 0.5687, "judge_b": 0.7208, "score": 0.6377, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 32, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A developer's llama-sleep wrapper for llama-swap parks idle llama.cpp models' GPU memory in host RAM instead of restarting them, cutting a 10.3-second model swap on a single RTX 3060 to 2.1 seconds while keeping cached prompts.", "page": "https://rubric.hrih.si/editions/2026-11-07/42/", "card": "https://rubric.hrih.si/og/2026-11-07/42.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.495, "judge_a": 0.4, "judge_b": 0.58, "mean": 0.4917, "spread": 0.18 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.72, "judge_b": 0.9, "mean": 0.7733, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.417, "judge_a": 0.4, "judge_b": 0.4, "mean": 0.4056, "spread": 0.017 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.78, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.72, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.987, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8889, "spread": 0.187 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 1, "judge_a": 0.85, "judge_b": 0.91, "mean": 0.92, "spread": 0.15 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.78, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.8867, "spread": 0.2 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0433, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.1733, "spread": 0.21 } } }, { "rank": 43, "id": "st_20261107_16838", "title": "Making musl and proprietary NVIDIA drivers work together", "url": "https://chimera-linux.org/news/2026/11/musl-nvidia.html", "domain": "chimera-linux.org", "site": "Chimera Linux", "categories": [ "programming", "hardware" ], "category_names": [ "Programming & dev tools", "Hardware & devices" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T13:23:14.000Z", "jev": 0.5193, "judge_a": 0.5478, "judge_b": 0.6639, "score": 0.5769, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 78, "topic_penalty": 0.02, "repeat_penalty": 0, "summary": "A developer documents making NVIDIA's proprietary Linux drivers, which are built against glibc, run on musl-based distributions such as Chimera Linux. The work covers switching to GLVND for vendor-neutral OpenGL dispatch, patching the gcompat shim to supply missing glibc symbols, and using patchelf to sanitize driver libraries, with the remaining…", "page": "https://rubric.hrih.si/editions/2026-11-07/43/", "card": "https://rubric.hrih.si/og/2026-11-07/43.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.498, "judge_a": 0.35, "judge_b": 0.55, "mean": 0.4658, "spread": 0.2 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.51, "judge_a": 0.7, "judge_b": 0.8, "mean": 0.67, "spread": 0.29 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.347, "judge_a": 0.4, "judge_b": 0.38, "mean": 0.3756, "spread": 0.053 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.535, "judge_a": 0.55, "judge_b": 0.68, "mean": 0.5883, "spread": 0.145 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.86, "judge_a": 0.8, "judge_b": 0.82, "mean": 0.8267, "spread": 0.06 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.92, "judge_b": 0.95, "mean": 0.9558, "spread": 0.078 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9133, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.03, "judge_b": 0.02, "mean": 0.0333, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.1433, "spread": 0.12 } } }, { "rank": 44, "id": "st_20261107_16692", "title": "UK ministers seek powers to investigate when AI agents are let off the leash", "url": "https://www.theregister.com/ai-and-ml/2026/11/06/uk-ministers-seek-powers-to-investigate-when-ai-agents-are-let-off-the-leash/5307889", "domain": "theregister.com", "site": "The Register", "categories": [ "ai", "security", "policy" ], "category_names": [ "AI & machine learning", "Security", "Policy & regulation" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T11:00:00.000Z", "jev": 0.5569, "judge_a": 0.4385, "judge_b": 0.6753, "score": 0.5569, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 99, "topic_penalty": 0, "repeat_penalty": 0, "summary": "UK ministers are seeking powers, through an amendment to the Cyber Security and Resilience Bill, to regulate how essential service operators and digital providers deploy agentic AI and report when those systems lose control.", "page": "https://rubric.hrih.si/editions/2026-11-07/44/", "card": "https://rubric.hrih.si/og/2026-11-07/44.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.648, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.6225, "spread": 0.22 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.58, "judge_a": 0.55, "judge_b": 0.82, "mean": 0.65, "spread": 0.27 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.69, "judge_a": 0.5, "judge_b": 0.61, "mean": 0.6, "spread": 0.19 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.53, "judge_a": 0.4, "judge_b": 0.62, "mean": 0.5167, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.383, "judge_a": 0.45, "judge_b": 0.69, "mean": 0.5078, "spread": 0.307 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.413, "judge_a": 0.15, "judge_b": 0.28, "mean": 0.2808, "spread": 0.262 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.7, "judge_b": 0.88, "mean": 0.8067, "spread": 0.18 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.05, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.22, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.2133, "spread": 0.18 } } }, { "rank": 45, "id": "st_20261107_16609", "title": "The Secret Leak Test: a prompt injection filter vs. attacks", "url": "https://hodor.gatebolt.com/blog/secret-leak-test", "domain": "hodor.gatebolt.com", "site": "Hodor", "categories": [ "ai", "security" ], "category_names": [ "AI & machine learning", "Security" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T08:32:59.000Z", "jev": 0.671, "judge_a": 0.4835, "judge_b": 0.7536, "score": 0.6359, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 33, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "GateBolt's author tested its message-and-reply filter against prompt injection, using a chatbot holding a password and public attack datasets plus AI-run multi-turn battles.", "page": "https://rubric.hrih.si/editions/2026-11-07/45/", "card": "https://rubric.hrih.si/og/2026-11-07/45.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.588, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5558, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.75, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7833, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.567, "judge_a": 0.55, "judge_b": 0.65, "mean": 0.5889, "spread": 0.1 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.733, "judge_a": 0.65, "judge_b": 0.82, "mean": 0.7342, "spread": 0.17 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.997, "judge_a": 0.7, "judge_b": 0.88, "mean": 0.8589, "spread": 0.297 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.99, "judge_a": 0.75, "judge_b": 0.82, "mean": 0.8533, "spread": 0.24 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.81, "judge_a": 0.85, "judge_b": 0.99, "mean": 0.8833, "spread": 0.18 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.45, "judge_b": 0.18, "mean": 0.2433, "spread": 0.35 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.19, "judge_a": 0.35, "judge_b": 0.08, "mean": 0.2067, "spread": 0.27 } } }, { "rank": 46, "id": "st_20261107_17430", "title": "The Port That Does Not Exist", "url": "https://acid.vegas/blog/the-port-that-does-not-exist/", "domain": "acid.vegas", "site": "acidvegas", "categories": [ "security", "it", "hardware" ], "category_names": [ "Security", "IT & networking", "Hardware & devices" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T22:52:46.000Z", "jev": 0.6455, "judge_a": 0.5065, "judge_b": 0.7543, "score": 0.6354, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 34, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A blog post by ACIDVEGAS reports that the author ran an IRC network on TCP port 0 and then masscanned all of IPv4 for open port-0 listeners.", "page": "https://rubric.hrih.si/editions/2026-11-07/46/", "card": "https://rubric.hrih.si/og/2026-11-07/46.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.505, "judge_a": 0.35, "judge_b": 0.72, "mean": 0.525, "spread": 0.37 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.76, "judge_a": 0.7, "judge_b": 0.93, "mean": 0.7967, "spread": 0.23 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.72, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.6133, "spread": 0.22 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.588, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5558, "spread": 0.28 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.97, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8833, "spread": 0.17 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.9, "judge_b": 0.91, "mean": 0.9358, "spread": 0.098 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.92, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.03, "judge_a": 0.05, "judge_b": 0.06, "mean": 0.0467, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.3, "judge_a": 0.3, "judge_b": 0.16, "mean": 0.2533, "spread": 0.14 } } }, { "rank": 47, "id": "st_20261107_17359", "title": "PyTorch 2.15 tackles the 'one bad GPU' problem", "url": "https://www.theregister.com/devops/2026/11/06/pytorch-215-tackles-the-one-bad-gpu-problem/5308227", "domain": "theregister.com", "site": "The Register", "categories": [ "ai", "devops" ], "category_names": [ "AI & machine learning", "Infrastructure & DevOps" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T18:36:42.000Z", "jev": 0.57, "judge_a": 0.594, "judge_b": 0.7298, "score": 0.6312, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 35, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "PyTorch 2.15 rewrites its NCCL backend, the layer that handles GPU-to-GPU communication, to address \"one bad GPU\" failures in training clusters. The new design allows one communicator per process group, a faulty or slow GPU can be excluded and replaced by an external orchestrator, and a \"flight recorder\" aids diagnosis.…", "page": "https://rubric.hrih.si/editions/2026-11-07/47/", "card": "https://rubric.hrih.si/og/2026-11-07/47.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.693, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6542, "spread": 0.17 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.64, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.7333, "spread": 0.22 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.537, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.6056, "spread": 0.143 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.688, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7358, "spread": 0.132 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.303, "judge_a": 0.5, "judge_b": 0.52, "mean": 0.4411, "spread": 0.217 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.713, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.6442, "spread": 0.113 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.9, "judge_b": 0.94, "mean": 0.8933, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.1, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.49, "judge_a": 0.3, "judge_b": 0.22, "mean": 0.3367, "spread": 0.27 } } }, { "rank": 48, "id": "st_20261107_17199", "title": "Project Eve: Claude Plays MMORPGs", "url": "https://www.anthropic.com/research/project-eve-claude-plays-mmorpgs", "domain": "anthropic.com", "site": "anthropic.com", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T18:07:03.000Z", "jev": 0.6228, "judge_a": 0.447, "judge_b": 0.8232, "score": 0.6309, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 36, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Anthropic's Fellows Program study tested Claude models as traders in EVE Online, a player-driven sci-fi MMORPG whose market mirrors real exchange mechanics. Using a simulator replaying actual 2024–2026 order books, newer Claude models grew capital steadily while rule-based strategies lost money, mostly through low-risk spread capture.", "page": "https://rubric.hrih.si/editions/2026-11-07/48/", "card": "https://rubric.hrih.si/og/2026-11-07/48.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.625, "judge_a": 0.4, "judge_b": 0.82, "mean": 0.615, "spread": 0.42 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.68, "judge_a": 0.7, "judge_b": 0.94, "mean": 0.7733, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.647, "judge_a": 0.35, "judge_b": 0.78, "mean": 0.5922, "spread": 0.43 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.543, "judge_a": 0.35, "judge_b": 0.8, "mean": 0.5642, "spread": 0.45 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.907, "judge_a": 0.65, "judge_b": 0.9, "mean": 0.8189, "spread": 0.257 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.778, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6858, "spread": 0.28 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9167, "spread": 0.11 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.1, "spread": 0.08 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.25, "judge_b": 0.12, "mean": 0.19, "spread": 0.13 } } }, { "rank": 49, "id": "st_20261107_17432", "title": "Inkling 2 Small: 17A-594B trained in 68 hours on 12T tokens, beating Muse Spark", "url": "https://latent.space/p/ainews-inkling-2-small-17a-594b-trained-in-68-hours-on-12t-tokens-beating-muse-spark/", "domain": "latent.space", "site": "Latent Space", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T22:47:33.000Z", "jev": 0.6076, "judge_a": 0.549, "judge_b": 0.7296, "score": 0.6287, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 37, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Thinking Machines released Inkling 2 Small, a 594B-total, 17B-active multimodal mixture-of-experts model, available now on its Tinker platform, with Apache 2.0 weights promised for the following week after external safety testing.", "page": "https://rubric.hrih.si/editions/2026-11-07/49/", "card": "https://rubric.hrih.si/og/2026-11-07/49.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.725, "judge_a": 0.55, "judge_b": 0.82, "mean": 0.6983, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.68, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.7933, "spread": 0.22 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.803, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7744, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.678, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6658, "spread": 0.12 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.56, "judge_a": 0.45, "judge_b": 0.75, "mean": 0.5867, "spread": 0.3 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.815, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7317, "spread": 0.215 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.5, "judge_b": 0.93, "mean": 0.78, "spread": 0.43 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.15, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.1933, "spread": 0.1 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.75, "judge_a": 0.4, "judge_b": 0.5, "mean": 0.55, "spread": 0.35 } } }, { "rank": 50, "id": "st_20261107_16555", "title": "Active Benchmarking", "url": "https://www.brendangregg.com/activebenchmarking.html", "domain": "brendangregg.com", "site": "brendangregg.com", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T07:03:16.000Z", "jev": 0.5885, "judge_a": 0.521, "judge_b": 0.716, "score": 0.6085, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 51, "topic_penalty": 0.06, "repeat_penalty": 0, "summary": "Brendan Gregg's essay on active benchmarking argues that many benchmarks are run \"passively,\" with results collected and reported without checking what they actually measure, leading to misleading conclusions that can drive costly platform and architecture decisions.", "page": "https://rubric.hrih.si/editions/2026-11-07/50/", "card": "https://rubric.hrih.si/og/2026-11-07/50.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.525, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.4817, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.51, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.6433, "spread": 0.31 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.733, "judge_a": 0.6, "judge_b": 0.67, "mean": 0.6678, "spread": 0.133 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.778, "judge_a": 0.7, "judge_b": 0.88, "mean": 0.7858, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.367, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5122, "spread": 0.353 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.765, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.6483, "spread": 0.265 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.5, "mean": 0.76, "spread": 0.4 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.0467, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.2, "judge_b": 0.18, "mean": 0.1933, "spread": 0.02 } } }, { "rank": 51, "id": "st_20261107_17365", "title": "Hacking the meta ray-ban smartglasses", "url": "https://p0pcycle.com/2026/11/06/hacking-the-meta-ray-ban-smartglasses/", "domain": "p0pcycle.com", "site": "pamoutaf", "categories": [ "security", "hardware" ], "category_names": [ "Security", "Hardware & devices" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T20:50:24.000Z", "jev": 0.4714, "judge_a": 0.6498, "judge_b": 0.7632, "score": 0.6281, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 38, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A security researcher describes gaining arbitrary code execution on Meta's Ray-Ban smartglasses by pulling full firmware via the companion app's OTA update path, then exploiting the lack of signature verification on the Device Access Toolkit app to install a self-signed app and open a bind shell.", "page": "https://rubric.hrih.si/editions/2026-11-07/51/", "card": "https://rubric.hrih.si/og/2026-11-07/51.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.49, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.6067, "spread": 0.29 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.52, "judge_a": 0.9, "judge_b": 0.94, "mean": 0.7867, "spread": 0.42 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.303, "judge_a": 0.55, "judge_b": 0.42, "mean": 0.4244, "spread": 0.247 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.51, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.5767, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.48, "judge_a": 0.85, "judge_b": 0.86, "mean": 0.73, "spread": 0.38 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.898, "judge_a": 0.85, "judge_b": 0.9, "mean": 0.8825, "spread": 0.05 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.9, "judge_b": 0.99, "mean": 0.93, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.03, "judge_b": 0.03, "mean": 0.0367, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.34, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.22, "spread": 0.22 } } }, { "rank": 52, "id": "st_20261107_17413", "title": "Agents reveal secrets when offered to team up with each other", "url": "https://www.theregister.com/security/2026/11/06/agents-reveal-secrets-when-offered-to-team-up-with-each-other/5308260", "domain": "theregister.com", "site": "The Register", "categories": [ "ai", "security" ], "category_names": [ "AI & machine learning", "Security" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T21:39:23.000Z", "jev": 0.6694, "judge_a": 0.5, "judge_b": 0.7033, "score": 0.6242, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 39, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "In a hackathon test run by AE Studio, sponsored by the AI Alignment Foundation and reported by The Register, nine AI coding agents given a private token and a broken-pricing-service bug were offered help from a 24-agent forum.", "page": "https://rubric.hrih.si/editions/2026-11-07/52/", "card": "https://rubric.hrih.si/og/2026-11-07/52.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.705, "judge_a": 0.4, "judge_b": 0.72, "mean": 0.6083, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.79, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7967, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.85, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7433, "spread": 0.25 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.713, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6975, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.517, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5322, "spread": 0.28 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.558, "judge_a": 0.4, "judge_b": 0.48, "mean": 0.4792, "spread": 0.157 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.8, "judge_a": 0.7, "judge_b": 0.92, "mean": 0.8067, "spread": 0.22 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.1, "judge_b": 0.03, "mean": 0.0767, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.43, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5033, "spread": 0.28 } } }, { "rank": 53, "id": "st_20261107_16633", "title": "AI for scientific discovery: How researchers use Claude", "url": "https://www.anthropic.com/research/how-researchers-use-claude", "domain": "anthropic.com", "site": "anthropic.com", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T09:13:30.000Z", "jev": 0.5559, "judge_a": 0.572, "judge_b": 0.7236, "score": 0.6171, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 41, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Anthropic researchers analyzed 207,000 Claude and Claude Code sessions, using privacy-preserving classifiers, to map how scientists use the tools across fields and tasks.", "page": "https://rubric.hrih.si/editions/2026-11-07/53/", "card": "https://rubric.hrih.si/og/2026-11-07/53.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.64, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6733, "spread": 0.18 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.55, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.75, "spread": 0.35 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.583, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.6678, "spread": 0.237 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.473, "judge_a": 0.45, "judge_b": 0.65, "mean": 0.5242, "spread": 0.2 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.947, "judge_a": 0.8, "judge_b": 0.82, "mean": 0.8556, "spread": 0.147 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.735, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6717, "spread": 0.28 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.9167, "spread": 0.07 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.15, "judge_b": 0.28, "mean": 0.1833, "spread": 0.16 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.24, "judge_a": 0.35, "judge_b": 0.2, "mean": 0.2633, "spread": 0.15 } } }, { "rank": 54, "id": "st_20261107_16629", "title": "Treeverse/Dvc: Data Versioning and ML Experiments", "url": "https://github.com/treeverse/dvc", "domain": "github.com", "site": "GitHub", "categories": [ "programming", "ai", "data" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Data" ], "format": "Other", "depth": "three_judges", "published_at": "2026-11-06T09:00:46.000Z", "jev": 0.5266, "judge_a": 0.5335, "judge_b": 0.6077, "score": 0.5559, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 100, "topic_penalty": 0.02, "repeat_penalty": 0, "summary": "DVC (Data Version Control) is an open-source command-line tool and VS Code extension from Iterative for building reproducible machine learning projects. It versions large data and models in cloud or network storage while keeping version references in Git, and tracks experiments locally without a server.", "page": "https://rubric.hrih.si/editions/2026-11-07/54/", "card": "https://rubric.hrih.si/og/2026-11-07/54.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.618, "judge_a": 0.45, "judge_b": 0.55, "mean": 0.5392, "spread": 0.168 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.34, "judge_a": 0.7, "judge_b": 0.65, "mean": 0.5633, "spread": 0.36 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.487, "judge_a": 0.6, "judge_b": 0.55, "mean": 0.5456, "spread": 0.113 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.775, "judge_a": 0.8, "judge_b": 0.78, "mean": 0.785, "spread": 0.025 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.65, "judge_a": 0.3, "judge_b": 0.78, "mean": 0.5767, "spread": 0.48 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.77, "judge_a": 0.55, "judge_b": 0.62, "mean": 0.6467, "spread": 0.22 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.9, "judge_b": 0.82, "mean": 0.87, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.22, "judge_a": 0.4, "judge_b": 0.18, "mean": 0.2667, "spread": 0.22 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.16, "spread": 0.08 } } }, { "rank": 55, "id": "st_20261107_17117", "title": "Anthropic IPO Valuation Talks Drop to $1.3T as Major Investors Balk", "url": "https://epyhia.com/anthropic-ipo-valuation-talks-drop-to-1-3-trillion-as-major-investors", "domain": "epyhia.com", "site": "epyhia.com", "categories": [ "ai", "startups" ], "category_names": [ "AI & machine learning", "Startups & founders" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T17:21:15.000Z", "jev": 0.5111, "judge_a": 0.5641, "judge_b": 0.7718, "score": 0.6156, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 42, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Bloomberg and the Wall Street Journal report that Anthropic's planned IPO valuation talks have fallen from $2 trillion in early October to $1.2–1.3 trillion, after prospective investors balked following a 6.4% Nasdaq 100 plunge. Investors are still working toward an offering, though its timing is uncertain.", "page": "https://rubric.hrih.si/editions/2026-11-07/55/", "card": "https://rubric.hrih.si/og/2026-11-07/55.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.675, "judge_a": 0.6, "judge_b": 0.88, "mean": 0.7183, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.62, "judge_a": 0.85, "judge_b": 0.95, "mean": 0.8067, "spread": 0.33 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.577, "judge_a": 0.67, "judge_b": 0.82, "mean": 0.6889, "spread": 0.243 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.293, "judge_a": 0.35, "judge_b": 0.62, "mean": 0.4208, "spread": 0.328 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.64, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.6067, "spread": 0.18 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.383, "judge_a": 0.2, "judge_b": 0.36, "mean": 0.3142, "spread": 0.183 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.85, "judge_a": 0.85, "judge_b": 0.92, "mean": 0.8733, "spread": 0.07 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.05, "judge_b": 0.04, "mean": 0.0533, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.2567, "spread": 0.12 } } }, { "rank": 56, "id": "st_20261107_16836", "title": "DigiCert admits 16-month security lapse exposed customer information", "url": "https://www.theregister.com/security/2026/11/06/digicert-admits-16-month-security-lapse-exposed-customer-information/5308097", "domain": "theregister.com", "site": "The Register", "categories": [ "security" ], "category_names": [ "Security" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T13:15:00.000Z", "jev": 0.5224, "judge_a": 0.607, "judge_b": 0.713, "score": 0.6141, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 43, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "DigiCert, a certificate authority, says a February 2025 server migration left an internal archive exposed to the public internet for over 16 months, letting an unauthorized user reach customer records including names, contact details, and possibly passports and driver's licenses.", "page": "https://rubric.hrih.si/editions/2026-11-07/56/", "card": "https://rubric.hrih.si/og/2026-11-07/56.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.613, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6275, "spread": 0.17 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.67, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.7833, "spread": 0.21 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.197, "judge_a": 0.55, "judge_b": 0.55, "mean": 0.4322, "spread": 0.353 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.543, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.6075, "spread": 0.138 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.493, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.5711, "spread": 0.127 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.64, "judge_a": 0.4, "judge_b": 0.43, "mean": 0.49, "spread": 0.24 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.9, "judge_b": 0.97, "mean": 0.9133, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.0433, "spread": 0.06 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.28, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.1867, "spread": 0.2 } } }, { "rank": 57, "id": "st_20261107_16742", "title": "Apple will soon scan iCloud links for “Harmful Content”", "url": "https://www.heise.de/en/news/Apple-will-soon-scan-iCloud-links-for-Harmful-Content-11515271.html", "domain": "heise.de", "site": "heise online English", "categories": [ "security", "policy" ], "category_names": [ "Security", "Policy & regulation" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T11:36:00.000Z", "jev": 0.5271, "judge_a": 0.547, "judge_b": 0.5877, "score": 0.5539, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 102, "topic_penalty": 0.02, "repeat_penalty": 0, "summary": "Apple plans to scan iCloud content that users share via \"Anyone with the Link\" for \"Harmful Content,\" according to a support document published alongside iOS 27.3 developer betas. Such content would no longer be protected by end-to-end encryption or Advanced Data Protection.", "page": "https://rubric.hrih.si/editions/2026-11-07/57/", "card": "https://rubric.hrih.si/og/2026-11-07/57.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.663, "judge_a": 0.55, "judge_b": 0.58, "mean": 0.5975, "spread": 0.112 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.7, "judge_b": 0.72, "mean": 0.69, "spread": 0.07 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.323, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.5144, "spread": 0.297 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.45, "judge_a": 0.5, "judge_b": 0.48, "mean": 0.4767, "spread": 0.05 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.65, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6567, "spread": 0.12 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.59, "judge_a": 0.35, "judge_b": 0.22, "mean": 0.3867, "spread": 0.37 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.8, "judge_b": 0.94, "mean": 0.88, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0567, "spread": 0.06 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.29, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.2367, "spread": 0.18 } } }, { "rank": 58, "id": "st_20261107_17271", "title": "What Huawei’s Horinger data center reveals about China’s compute gap", "url": "https://epoch.ai/publications/huawei-horinger-china-compute-gap", "domain": "epoch.ai", "site": "Epoch AI", "categories": [ "ai", "hardware" ], "category_names": [ "AI & machine learning", "Hardware & devices" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T00:00:00.000Z", "jev": 0.5476, "judge_a": 0.4873, "judge_b": 0.7996, "score": 0.6114, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 46, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "An analysis by Epoch AI estimates that Huawei's Horinger AI data center in Inner Mongolia has roughly 286 MW of IT power capacity across four operational buildings, mostly housing Ascend 910C chips.", "page": "https://rubric.hrih.si/editions/2026-11-07/58/", "card": "https://rubric.hrih.si/og/2026-11-07/58.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.608, "judge_a": 0.4, "judge_b": 0.82, "mean": 0.6092, "spread": 0.42 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.61, "judge_a": 0.7, "judge_b": 0.92, "mean": 0.7433, "spread": 0.31 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.587, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6222, "spread": 0.28 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.328, "judge_a": 0.3, "judge_b": 0.68, "mean": 0.4358, "spread": 0.38 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.87, "judge_a": 0.72, "judge_b": 0.88, "mean": 0.8233, "spread": 0.16 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.9, "judge_a": 0.6, "judge_b": 0.87, "mean": 0.79, "spread": 0.3 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.8, "judge_b": 0.98, "mean": 0.8933, "spread": 0.18 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.03, "judge_b": 0.01, "mean": 0.0333, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.25, "judge_a": 0.2, "judge_b": 0.2, "mean": 0.2167, "spread": 0.05 } } }, { "rank": 59, "id": "st_20261107_16422", "title": "Threat actor targets Indian fintechs on weekends when defenders aren’t watching", "url": "https://www.theregister.com/security/2026/11/06/threat-actor-targets-indian-fintechs-on-weekends-when-defenders-arent-watching/5308000", "domain": "theregister.com", "site": "The Register", "categories": [ "security", "fintech" ], "category_names": [ "Security", "Fintech" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T01:17:40.000Z", "jev": 0.5675, "judge_a": 0.5808, "judge_b": 0.6855, "score": 0.6112, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 47, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "India's CERT-IN has warned fintechs, payment providers and lenders of escalating attacks by a threat actor it calls CRAFTY FALCON, which it describes as AI-assisted. The group reportedly re-targets victims after incomplete remediation and times attacks to weekends and holidays, when monitoring is thin.", "page": "https://rubric.hrih.si/editions/2026-11-07/59/", "card": "https://rubric.hrih.si/og/2026-11-07/59.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.56, "judge_a": 0.5, "judge_b": 0.65, "mean": 0.57, "spread": 0.15 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.71, "judge_a": 0.7, "judge_b": 0.85, "mean": 0.7533, "spread": 0.15 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.447, "judge_a": 0.5, "judge_b": 0.55, "mean": 0.4989, "spread": 0.103 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.638, "judge_a": 0.7, "judge_b": 0.7, "mean": 0.6792, "spread": 0.063 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.443, "judge_a": 0.45, "judge_b": 0.62, "mean": 0.5044, "spread": 0.177 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.673, "judge_a": 0.45, "judge_b": 0.48, "mean": 0.5342, "spread": 0.222 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.85, "judge_b": 0.9, "mean": 0.8733, "spread": 0.05 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.03, "judge_b": 0.02, "mean": 0.0433, "spread": 0.06 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.31, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.21, "spread": 0.19 } } }, { "rank": 60, "id": "st_20261107_17312", "title": "Can LLMs redesign themselves as they train?", "url": "https://www.doubleai.com/research/can-llms-redesign-themselves-as-they-train", "domain": "doubleai.com", "site": "doubleAI", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T20:30:38.000Z", "jev": 0.6085, "judge_a": 0.475, "judge_b": 0.7487, "score": 0.6107, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 48, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "An article on arXiv-style research describes letting coding agents redesign a neural network's architecture during training, by inserting code-written layers and using a \"phantom gradient\" to find where the model needs help.", "page": "https://rubric.hrih.si/editions/2026-11-07/60/", "card": "https://rubric.hrih.si/og/2026-11-07/60.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.665, "judge_a": 0.55, "judge_b": 0.86, "mean": 0.6917, "spread": 0.31 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.72, "judge_a": 0.7, "judge_b": 0.95, "mean": 0.79, "spread": 0.25 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.717, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6789, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.555, "judge_a": 0.35, "judge_b": 0.76, "mean": 0.555, "spread": 0.41 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.753, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6778, "spread": 0.28 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.933, "judge_a": 0.8, "judge_b": 0.96, "mean": 0.8975, "spread": 0.16 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.99, "mean": 0.9233, "spread": 0.11 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.13, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.15, "spread": 0.08 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.52, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6133, "spread": 0.2 } } }, { "rank": 61, "id": "st_20261107_16447", "title": "Closed AI models dominate the price-performance frontier", "url": "https://village.dev/blog/closed-ai-models-now-dominate-the-price-performance-frontier/", "domain": "village.dev", "site": "Village", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T02:49:56.000Z", "jev": 0.6047, "judge_a": 0.417, "judge_b": 0.8058, "score": 0.6091, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 50, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Village's LLM Atlas, which maps the strongest AI model at each price point, finds that open-weight models now hold the best value at just 3% of budgets, down from 41% in June. OpenAI's GPT-6 Luna and GPT-6.1 Terra and Anthropic's Claude Haiku 5.5 took back most of that ground.", "page": "https://rubric.hrih.si/editions/2026-11-07/61/", "card": "https://rubric.hrih.si/og/2026-11-07/61.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.635, "judge_a": 0.35, "judge_b": 0.82, "mean": 0.6017, "spread": 0.47 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.55, "judge_b": 0.9, "mean": 0.7, "spread": 0.35 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.753, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.7044, "spread": 0.36 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.578, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6192, "spread": 0.28 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.75, "judge_a": 0.5, "judge_b": 0.74, "mean": 0.6633, "spread": 0.25 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.68, "judge_a": 0.35, "judge_b": 0.68, "mean": 0.57, "spread": 0.33 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.6, "judge_b": 0.98, "mean": 0.8067, "spread": 0.38 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.0933, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.33, "judge_a": 0.35, "judge_b": 0.12, "mean": 0.2667, "spread": 0.23 } } }, { "rank": 62, "id": "st_20261107_16856", "title": "Together Compiler", "url": "https://www.alphaxiv.org/abs/2611.together-compiler-continual-ai-inference", "domain": "alphaxiv.org", "site": "alphaXiv", "categories": [ "ai", "devops" ], "category_names": [ "AI & machine learning", "Infrastructure & DevOps" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T13:34:27.000Z", "jev": 0.5271, "judge_a": 0.53, "judge_b": 0.7773, "score": 0.6114, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 45, "topic_penalty": 0.08, "repeat_penalty": 0.0051, "summary": "Together Compiler (TKCC) is an agent-driven system for optimizing LLM inference. Agents propose replacement forward-pass code tested on a persistent, already-loaded engine, screened by a fast tier and checked by a slower, more representative tier, with a cost model choosing GPU parallelism.", "page": "https://rubric.hrih.si/editions/2026-11-07/62/", "card": "https://rubric.hrih.si/og/2026-11-07/62.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.543, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.5875, "spread": 0.22 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.6, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7333, "spread": 0.3 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.457, "judge_a": 0.5, "judge_b": 0.58, "mean": 0.5122, "spread": 0.123 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.593, "judge_a": 0.6, "judge_b": 0.8, "mean": 0.6642, "spread": 0.208 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.637, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.6656, "spread": 0.36 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.815, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8317, "spread": 0.08 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.85, "judge_b": 0.97, "mean": 0.8967, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.2, "judge_b": 0.04, "mean": 0.1133, "spread": 0.16 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.42, "judge_a": 0.35, "judge_b": 0.08, "mean": 0.2833, "spread": 0.34 } } }, { "rank": 63, "id": "st_20261107_17332", "title": "Discord is rebuilding in Rust, leaving React and Electron behind", "url": "https://www.heise.de/en/news/Discord-is-rebuilding-in-Rust-leaving-React-and-Electron-behind-11516113.html", "domain": "heise.de", "site": "heise online English", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T20:23:00.000Z", "jev": 0.54, "judge_a": 0.5791, "judge_b": 0.697, "score": 0.6053, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 53, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Discord, per its CTO Stanislav Vishnevskiy's announcement on X, is replacing Electron and React in its clients with a Rust core, the GPUI framework for the interface, and native Android and iOS UIs sharing Rust logic in place of React Native.", "page": "https://rubric.hrih.si/editions/2026-11-07/63/", "card": "https://rubric.hrih.si/og/2026-11-07/63.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.648, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.6425, "spread": 0.08 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.68, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.7467, "spread": 0.18 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.327, "judge_a": 0.67, "judge_b": 0.72, "mean": 0.5722, "spread": 0.393 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.578, "judge_a": 0.55, "judge_b": 0.7, "mean": 0.6092, "spread": 0.15 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.457, "judge_a": 0.4, "judge_b": 0.72, "mean": 0.5256, "spread": 0.32 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.643, "judge_a": 0.45, "judge_b": 0.56, "mean": 0.5508, "spread": 0.192 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9133, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.05, "judge_b": 0.08, "mean": 0.0733, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.39, "judge_a": 0.3, "judge_b": 0.32, "mean": 0.3367, "spread": 0.09 } } }, { "rank": 64, "id": "st_20261107_16824", "title": "EmbeddingGemma 2: Unified and Efficient Multimodal Representations", "url": "https://arxiv.org/abs/2610.37364", "domain": "arxiv.org", "site": "arXiv", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T13:06:35.000Z", "jev": 0.673, "judge_a": 0.6175, "judge_b": 0.8249, "score": 0.6051, "contested": false, "flag_penalty": 0, "freshness_penalty": 0.1, "score_rank": 54, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "EmbeddingGemma 2 is an open 744M-parameter embedding model from Google that maps text, images, video, audio, and interleaved combinations into one shared vector space. Its modality encoders are modular, so it can run as text-only (271M), vision-language (439M), or audio-language (577M).", "page": "https://rubric.hrih.si/editions/2026-11-07/64/", "card": "https://rubric.hrih.si/og/2026-11-07/64.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.725, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.685, "spread": 0.23 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.66, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7533, "spread": 0.24 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.627, "judge_a": 0.55, "judge_b": 0.8, "mean": 0.6589, "spread": 0.25 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.753, "judge_a": 0.7, "judge_b": 0.84, "mean": 0.7642, "spread": 0.14 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.977, "judge_a": 0.85, "judge_b": 0.94, "mean": 0.9222, "spread": 0.127 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.988, "judge_a": 0.85, "judge_b": 0.98, "mean": 0.9392, "spread": 0.138 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.9, "judge_b": 0.99, "mean": 0.9267, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.0867, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.43, "judge_a": 0.3, "judge_b": 0.14, "mean": 0.29, "spread": 0.29 } } }, { "rank": 65, "id": "st_20261107_16475", "title": "What does LoRA specialized Muon optimizer even mean?", "url": "https://theblackcat102.github.io/do-lora-optimizers-matter/", "domain": "theblackcat102.github.io", "site": "theblackcat102", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T03:32:41.000Z", "jev": 0.5144, "judge_a": 0.537, "judge_b": 0.7623, "score": 0.6045, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 55, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A solo replication on Qwen3-1.7B with OpenThoughts data tested LoRA-specific optimizers against AdamW and Muon at ranks 16 to 256, using one seed. Anpaure's reverse-engineered Dimuon variant led at every rank, with the margin growing with rank.", "page": "https://rubric.hrih.si/editions/2026-11-07/65/", "card": "https://rubric.hrih.si/og/2026-11-07/65.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.502, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5275, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.55, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7167, "spread": 0.35 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.407, "judge_a": 0.3, "judge_b": 0.58, "mean": 0.4289, "spread": 0.28 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.618, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.6792, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.937, "judge_a": 0.75, "judge_b": 0.74, "mean": 0.8089, "spread": 0.197 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.85, "judge_b": 0.9, "mean": 0.9158, "spread": 0.148 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.92, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.05, "spread": 0.08 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.56, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.33, "spread": 0.38 } } }, { "rank": 66, "id": "st_20261107_16611", "title": "Practical guide on out of tree mitigations on Linux", "url": "https://clehaxze.tw/gemlog/2026/11-04-practical-guide-on-out-of-tree-mitigations-on-linux.gmi", "domain": "clehaxze.tw", "site": "Martin's website/blog thingy", "categories": [ "programming", "security", "devops", "it" ], "category_names": [ "Programming & dev tools", "Security", "Infrastructure & DevOps", "IT & networking" ], "format": "Tutorial", "depth": "three_judges", "published_at": "2026-11-06T08:53:27.000Z", "jev": 0.6571, "judge_a": 0.5418, "judge_b": 0.6101, "score": 0.603, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 56, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A practical guide explains how to add out-of-tree security mitigations to Linux services, written by a blogger responding to a reader question. It covers hardened allocators (hardened_malloc, mimalloc secure mode, Scudo), with a demo showing a heap overflow caught only under hardened allocators.", "page": "https://rubric.hrih.si/editions/2026-11-07/66/", "card": "https://rubric.hrih.si/og/2026-11-07/66.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.513, "judge_a": 0.3, "judge_b": 0.56, "mean": 0.4575, "spread": 0.26 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.56, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6267, "spread": 0.16 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.653, "judge_a": 0.5, "judge_b": 0.55, "mean": 0.5678, "spread": 0.153 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.925, "judge_a": 0.85, "judge_b": 0.78, "mean": 0.8517, "spread": 0.145 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.633, "judge_a": 0.45, "judge_b": 0.57, "mean": 0.5511, "spread": 0.183 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.85, "judge_b": 0.82, "mean": 0.8892, "spread": 0.178 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.6, "judge_b": 0.92, "mean": 0.8067, "spread": 0.32 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.03, "judge_a": 0.03, "judge_b": 0.01, "mean": 0.0233, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.19, "judge_a": 0.25, "judge_b": 0.62, "mean": 0.3533, "spread": 0.43 } } }, { "rank": 67, "id": "st_20261107_16882", "title": "Trustworthy AI", "url": "https://timmykokke.com/blog/2026/2026-11-04-trustworthy-ai/", "domain": "timmykokke.com", "site": "Timmy Kokke", "categories": [ "ai", "policy" ], "category_names": [ "AI & machine learning", "Policy & regulation" ], "format": "Analysis", "depth": "claim_checked", "published_at": "2026-11-06T14:04:50.000Z", "jev": 0.7701, "judge_a": 0.5715, "judge_b": 0.7603, "score": 0.6006, "contested": true, "flag_penalty": 0.1, "freshness_penalty": 0, "score_rank": 58, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "The post is a practical guide for software engineers on the EU AI Act, arguing that compliance starts with describing what an AI system is used for and who it affects, not with the model itself.", "page": "https://rubric.hrih.si/editions/2026-11-07/67/", "card": "https://rubric.hrih.si/og/2026-11-07/67.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.708, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6592, "spread": 0.17 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.73, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.877, "judge_a": 0.7, "judge_b": 0.68, "mean": 0.7522, "spread": 0.197 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.953, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.8108, "spread": 0.253 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.597, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.6722, "spread": 0.223 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.66, "judge_a": 0.35, "judge_b": 0.58, "mean": 0.53, "spread": 0.31 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.45, "judge_b": 0.93, "mean": 0.7633, "spread": 0.48 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.05, "judge_b": 0.02, "mean": 0.04, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.11, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.13, "spread": 0.12 } } }, { "rank": 68, "id": "st_20261107_17295", "title": "Fake security tools on GitHub push backdoors and infostealers", "url": "https://www.bleepingcomputer.com/news/security/fake-security-tools-on-github-push-backdoors-and-infostealers/", "domain": "bleepingcomputer.com", "site": "BleepingComputer", "categories": [ "security" ], "category_names": [ "Security" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T19:33:50.000Z", "jev": 0.6548, "judge_a": 0.4885, "judge_b": 0.6788, "score": 0.6073, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 52, "topic_penalty": 0.08, "repeat_penalty": 0.0073, "summary": "A threat actor has been publishing backdoored security tools on GitHub, including fake scanners and log-parsing scripts aimed at cybersecurity professionals. Dr. Web reports more than 2,000 Windows systems compromised, with payloads including the AsyncRAT backdoor, clipboard-swapping crypto malware, and an infostealer that targets browser data and Discord tokens.", "page": "https://rubric.hrih.si/editions/2026-11-07/68/", "card": "https://rubric.hrih.si/og/2026-11-07/68.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.593, "judge_a": 0.4, "judge_b": 0.63, "mean": 0.5408, "spread": 0.23 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.74, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.7333, "spread": 0.26 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.647, "judge_a": 0.5, "judge_b": 0.48, "mean": 0.5422, "spread": 0.167 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.705, "judge_a": 0.6, "judge_b": 0.7, "mean": 0.6683, "spread": 0.105 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.597, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6389, "spread": 0.123 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.81, "judge_a": 0.6, "judge_b": 0.66, "mean": 0.69, "spread": 0.21 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.3, "judge_b": 0.94, "mean": 0.7, "spread": 0.64 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.09, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.14, "judge_a": 0.15, "judge_b": 0.12, "mean": 0.1367, "spread": 0.03 } } }, { "rank": 69, "id": "st_20261107_16615", "title": "Dutch 'idealist' bank ASN blocks non-Google Android, including GrapheneOS", "url": "https://github.com/PrivSec-dev/banking-apps-compat-report/issues/105", "domain": "github.com", "site": "GitHub", "categories": [ "security" ], "category_names": [ "Security" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T09:27:12.000Z", "jev": 0.6081, "judge_a": 0.5513, "judge_b": 0.64, "score": 0.5998, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 59, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Key attestation certificates, as the Android spec describes them, carry the challenge, security level, a RootOfTrust extension (verifiedBootState, verifiedBootKey hash, deviceLocked) and attestationApplicationId, all signed by the chip and chaining to Google's roots.", "page": "https://rubric.hrih.si/editions/2026-11-07/69/", "card": "https://rubric.hrih.si/og/2026-11-07/69.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.525, "judge_a": 0.35, "judge_b": 0.58, "mean": 0.485, "spread": 0.23 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7267, "spread": 0.08 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.503, "judge_a": 0.5, "judge_b": 0.42, "mean": 0.4744, "spread": 0.083 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.775, "judge_a": 0.7, "judge_b": 0.72, "mean": 0.7317, "spread": 0.075 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.653, "judge_a": 0.6, "judge_b": 0.67, "mean": 0.6411, "spread": 0.07 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.95, "judge_a": 0.8, "judge_b": 0.76, "mean": 0.8367, "spread": 0.19 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.7, "judge_b": 1, "mean": 0.8667, "spread": 0.3 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.03, "judge_b": 0.03, "mean": 0.04, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.45, "judge_a": 0.3, "judge_b": 0.28, "mean": 0.3433, "spread": 0.17 } } }, { "rank": 70, "id": "st_20261107_17380", "title": "VFlash: Enabling System-1 VLMs for Physical AI", "url": "https://research.nvidia.com/labs/lpr/vflash/", "domain": "research.nvidia.com", "site": "research.nvidia.com", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T21:04:04.000Z", "jev": 0.6012, "judge_a": 0.431, "judge_b": 0.7655, "score": 0.5992, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 60, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "VFlash is a speculative decoding framework for Physical AI that lets a lightweight drafter predict complete visual primitives, such as grasp points, 2D/3D trajectories, and bounding boxes, in one parallel pass while a frozen vision-language model verifies or conditions on them.", "page": "https://rubric.hrih.si/editions/2026-11-07/70/", "card": "https://rubric.hrih.si/og/2026-11-07/70.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.66, "judge_a": 0.45, "judge_b": 0.78, "mean": 0.63, "spread": 0.33 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.63, "judge_a": 0.6, "judge_b": 0.93, "mean": 0.72, "spread": 0.33 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.58, "judge_a": 0.4, "judge_b": 0.58, "mean": 0.52, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.613, "judge_a": 0.55, "judge_b": 0.76, "mean": 0.6408, "spread": 0.21 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.913, "judge_a": 0.6, "judge_b": 0.83, "mean": 0.7811, "spread": 0.313 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.918, "judge_a": 0.7, "judge_b": 0.92, "mean": 0.8458, "spread": 0.22 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.92, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.35, "judge_b": 0.12, "mean": 0.1967, "spread": 0.23 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.4, "judge_a": 0.45, "judge_b": 0.18, "mean": 0.3433, "spread": 0.27 } } }, { "rank": 71, "id": "st_20261107_17489", "title": "OpenAI Secures $11B at $1.25T Valuation", "url": "https://epyhia.com/openai-secures-11-billion-at-1-25-trillion-valuation-well-short-of", "domain": "epyhia.com", "site": "epyhia.com", "categories": [ "ai", "startups" ], "category_names": [ "AI & machine learning", "Startups & founders" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T23:42:48.000Z", "jev": 0.4674, "judge_a": 0.5639, "judge_b": 0.7738, "score": 0.6017, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 57, "topic_penalty": 0.08, "repeat_penalty": 0.004, "summary": "OpenAI has reportedly secured about $11 billion in commitments for the first close of a funding round at a $1.25 trillion valuation, less than half the $30 billion it sought at $1.4 trillion.", "page": "https://rubric.hrih.si/editions/2026-11-07/71/", "card": "https://rubric.hrih.si/og/2026-11-07/71.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.648, "judge_a": 0.62, "judge_b": 0.82, "mean": 0.6958, "spread": 0.2 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.56, "judge_a": 0.8, "judge_b": 0.96, "mean": 0.7733, "spread": 0.4 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.463, "judge_a": 0.72, "judge_b": 0.88, "mean": 0.6878, "spread": 0.417 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.283, "judge_a": 0.28, "judge_b": 0.67, "mean": 0.4108, "spread": 0.39 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.493, "judge_a": 0.7, "judge_b": 0.62, "mean": 0.6044, "spread": 0.207 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.19, "judge_a": 0.08, "judge_b": 0.08, "mean": 0.1167, "spread": 0.11 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.85, "judge_a": 0.85, "judge_b": 0.99, "mean": 0.8967, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.04, "judge_b": 0.03, "mean": 0.05, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.16, "spread": 0.08 } } }, { "rank": 72, "id": "st_20261107_16957", "title": "Turning the HEIC decoder into a SHA-256 computer", "url": "https://www.irisc2.com/blog/heic-hashquine", "domain": "irisc2.com", "site": "irisc2.com", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T15:25:42.000Z", "jev": 0.5931, "judge_a": 0.572, "judge_b": 0.6275, "score": 0.5975, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 62, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A team turned the HEVC decoder inside Apple's ImageIO into a SHA-256 computer: they wrote a HEIC file whose 564 pictures run the full SHA-256 compression, and the final decoded image displays the file's own hash.", "page": "https://rubric.hrih.si/editions/2026-11-07/72/", "card": "https://rubric.hrih.si/og/2026-11-07/72.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.555, "judge_a": 0.6, "judge_b": 0.55, "mean": 0.5683, "spread": 0.05 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.85, "judge_b": 0.9, "mean": 0.8267, "spread": 0.17 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.503, "judge_a": 0.5, "judge_b": 0.35, "mean": 0.4511, "spread": 0.153 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.418, "judge_a": 0.2, "judge_b": 0.4, "mean": 0.3392, "spread": 0.217 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.94, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.88, "spread": 0.14 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.95, "judge_b": 1, "mean": 0.9825, "spread": 0.05 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.95, "judge_b": 0.95, "mean": 0.9267, "spread": 0.07 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0433, "spread": 0.02 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.17, "judge_a": 0.25, "judge_b": 0.08, "mean": 0.1667, "spread": 0.17 } } }, { "rank": 73, "id": "st_20261107_16924", "title": "What we learned by watching Claude read Zapier.com", "url": "https://zapier.com/blog/what-we-learned-watching-claude-read-zapier/", "domain": "zapier.com", "site": "zapier.com", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T14:38:15.000Z", "jev": 0.6524, "judge_a": 0.4895, "judge_b": 0.6497, "score": 0.5971, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 63, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Zapier's engineering team built Agent Arm Runner, a tool that runs identical Claude Code tasks against production and preview versions of a page while holding the URL constant, to test how content changes affect what agents return.", "page": "https://rubric.hrih.si/editions/2026-11-07/73/", "card": "https://rubric.hrih.si/og/2026-11-07/73.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.535, "judge_a": 0.35, "judge_b": 0.55, "mean": 0.4783, "spread": 0.2 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.71, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7433, "spread": 0.12 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.577, "judge_a": 0.5, "judge_b": 0.55, "mean": 0.5422, "spread": 0.077 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.753, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.6775, "spread": 0.152 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.89, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.74, "spread": 0.34 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.9, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.7067, "spread": 0.3 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8467, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.1333, "spread": 0.12 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.3, "judge_b": 0.08, "mean": 0.18, "spread": 0.22 } } }, { "rank": 74, "id": "st_20261107_16440", "title": "Claude was faster and cheaper. We took the scenic route anyway", "url": "https://blog.grainulation.org/verify-then-vote/", "domain": "blog.grainulation.org", "site": "Grainulation Blog", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T02:59:39.000Z", "jev": 0.5904, "judge_a": 0.4505, "judge_b": 0.7436, "score": 0.5948, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 64, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "The author describes \"Verify, Then Vote,\" a method for picking among answers from several small local language models by checking whether each cited source supports its answer, then applying a fixed rule. On 1,000 questions, accuracy rose from 77.1% under simple majority voting to 85.0%.", "page": "https://rubric.hrih.si/editions/2026-11-07/74/", "card": "https://rubric.hrih.si/og/2026-11-07/74.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.5, "judge_a": 0.3, "judge_b": 0.65, "mean": 0.4833, "spread": 0.35 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.63, "judge_a": 0.6, "judge_b": 0.92, "mean": 0.7167, "spread": 0.32 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.6, "judge_a": 0.4, "judge_b": 0.58, "mean": 0.5267, "spread": 0.2 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.68, "judge_a": 0.5, "judge_b": 0.7, "mean": 0.6267, "spread": 0.2 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.987, "judge_a": 0.7, "judge_b": 0.94, "mean": 0.8756, "spread": 0.287 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.9, "judge_a": 0.75, "judge_b": 0.78, "mean": 0.81, "spread": 0.15 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.8, "judge_a": 0.8, "judge_b": 0.95, "mean": 0.85, "spread": 0.15 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.1, "judge_b": 0.03, "mean": 0.0633, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.4, "judge_a": 0.3, "judge_b": 0.08, "mean": 0.26, "spread": 0.32 } } }, { "rank": 75, "id": "st_20261107_17073", "title": "Global Digital Policy Roundup: October 2026", "url": "https://techpolicy.press/global-digital-policy-roundup-october-2026", "domain": "techpolicy.press", "site": "Tech Policy Press", "categories": [ "ai", "policy" ], "category_names": [ "AI & machine learning", "Policy & regulation" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T16:04:36.000Z", "jev": 0.5547, "judge_a": 0.3785, "judge_b": 0.7266, "score": 0.5533, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 104, "topic_penalty": 0.04, "repeat_penalty": 0, "summary": "Digital Policy Alert's October 2026 roundup summarizes global digital policy changes across content moderation, AI, competition, and data governance, drawing on daily monitoring of over 60 jurisdictions.", "page": "https://rubric.hrih.si/editions/2026-11-07/75/", "card": "https://rubric.hrih.si/og/2026-11-07/75.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.6, "judge_a": 0.35, "judge_b": 0.72, "mean": 0.5567, "spread": 0.37 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.42, "judge_a": 0.45, "judge_b": 0.82, "mean": 0.5633, "spread": 0.4 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.853, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.7378, "spread": 0.36 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.53, "judge_a": 0.45, "judge_b": 0.68, "mean": 0.5533, "spread": 0.23 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.667, "judge_a": 0.6, "judge_b": 0.88, "mean": 0.7156, "spread": 0.28 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.353, "judge_a": 0.1, "judge_b": 0.28, "mean": 0.2442, "spread": 0.252 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.93, "judge_a": 0.85, "judge_b": 0.99, "mean": 0.9233, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.25, "judge_b": 0.12, "mean": 0.1433, "spread": 0.19 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.11, "judge_a": 0.15, "judge_b": 0.05, "mean": 0.1033, "spread": 0.1 } } }, { "rank": 76, "id": "st_20261107_17182", "title": "We stress-tested AI spend limits under bursts and outages", "url": "https://stipend-ai.com/blog/ai-spend-limits-tested/", "domain": "stipend-ai.com", "site": "Stipend", "categories": [ "ai", "devops" ], "category_names": [ "AI & machine learning", "Infrastructure & DevOps" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T18:10:14.000Z", "jev": 0.7143, "judge_a": 0.4465, "judge_b": 0.6065, "score": 0.589, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 66, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Peter Olayemi and Onatunde Oladimeji stress-tested two AI spend-limit tools, LiteLLM (open-source proxy) and Revenium (commercial platform), using gpt-4o-mini with $0.05 per-customer budgets in October 2026. Configured with Redis and fail-closed settings, LiteLLM held budgets to about 2.5% overshoot, even through outages.", "page": "https://rubric.hrih.si/editions/2026-11-07/76/", "card": "https://rubric.hrih.si/og/2026-11-07/76.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.56, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.4933, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.81, "judge_a": 0.6, "judge_b": 0.84, "mean": 0.75, "spread": 0.24 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.437, "judge_a": 0.4, "judge_b": 0.48, "mean": 0.4389, "spread": 0.08 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.953, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.8375, "spread": 0.253 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.98, "judge_a": 0.72, "judge_b": 0.82, "mean": 0.84, "spread": 0.26 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.78, "judge_b": 0.75, "mean": 0.8425, "spread": 0.248 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9133, "spread": 0.12 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.45, "judge_b": 0.62, "mean": 0.3867, "spread": 0.53 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.2, "judge_b": 0.2, "mean": 0.2, "spread": 0 } } }, { "rank": 77, "id": "st_20261107_16695", "title": "Lockdown Mode Bypassed in New iPhone Exploit", "url": "https://www.privacyguides.org/news/2026/11/06/lockdown-mode-bypassed-in-new-iphone-exploit/", "domain": "privacyguides.org", "site": "Privacy Guides", "categories": [ "security" ], "category_names": [ "Security" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T11:30:42.000Z", "jev": 0.5151, "judge_a": 0.5421, "judge_b": 0.7087, "score": 0.5886, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 68, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Security researchers at Calif posted a video showing they could compromise an iPhone 18 Pro with Lockdown Mode enabled, using a malicious PDF delivered via AirDrop, then persist malware through an update to iOS 27.2. This challenges Apple's claim that no Lockdown Mode device had been hacked with spyware.", "page": "https://rubric.hrih.si/editions/2026-11-07/77/", "card": "https://rubric.hrih.si/og/2026-11-07/77.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.62, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6467, "spread": 0.12 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.64, "judge_a": 0.8, "judge_b": 0.91, "mean": 0.7833, "spread": 0.27 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.547, "judge_a": 0.67, "judge_b": 0.73, "mean": 0.6489, "spread": 0.183 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.455, "judge_a": 0.45, "judge_b": 0.62, "mean": 0.5083, "spread": 0.17 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.447, "judge_a": 0.35, "judge_b": 0.48, "mean": 0.4256, "spread": 0.13 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.528, "judge_a": 0.2, "judge_b": 0.43, "mean": 0.3858, "spread": 0.327 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.6, "judge_b": 0.96, "mean": 0.8167, "spread": 0.36 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0667, "spread": 0.09 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.38, "judge_a": 0.4, "judge_b": 0.18, "mean": 0.32, "spread": 0.22 } } }, { "rank": 78, "id": "st_20261107_16417", "title": "FrontierPhysics: Evaluating Agents for End-to-End Physics Research", "url": "https://arxiv.org/abs/2611.06910", "domain": "arxiv.org", "site": "arXiv", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T01:37:41.000Z", "jev": 0.5631, "judge_a": 0.44, "judge_b": 0.7584, "score": 0.5871, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 69, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "FrontierPhysics is a benchmark of 56 end-to-end physics research tasks, written by 42 physicists from their own projects across nine domains. Agents must review literature, plan, run experiments, and produce a manuscript, graded by deterministic verifiers and expert rubrics.", "page": "https://rubric.hrih.si/editions/2026-11-07/78/", "card": "https://rubric.hrih.si/og/2026-11-07/78.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.663, "judge_a": 0.45, "judge_b": 0.78, "mean": 0.6308, "spread": 0.33 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.52, "judge_a": 0.55, "judge_b": 0.9, "mean": 0.6567, "spread": 0.38 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.597, "judge_a": 0.35, "judge_b": 0.62, "mean": 0.5222, "spread": 0.27 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.423, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5008, "spread": 0.28 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.967, "judge_a": 0.7, "judge_b": 0.84, "mean": 0.8356, "spread": 0.267 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.88, "judge_a": 0.55, "judge_b": 0.82, "mean": 0.75, "spread": 0.33 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.9, "judge_b": 0.99, "mean": 0.9333, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.1, "judge_b": 0.03, "mean": 0.0667, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.24, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.24, "spread": 0.12 } } }, { "rank": 79, "id": "st_20261107_17314", "title": "AI Text Watermarking May Not Be Able to Meet the E.U.'S New Rules", "url": "https://spectrum.ieee.org/ai-watermark-openai-textgrain-europe", "domain": "spectrum.ieee.org", "site": "IEEE Spectrum", "categories": [ "ai", "policy" ], "category_names": [ "AI & machine learning", "Policy & regulation" ], "format": "News", "depth": "three_judges", "published_at": "2026-11-06T20:24:46.000Z", "jev": 0.5497, "judge_a": 0.538, "judge_b": 0.6668, "score": 0.5848, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 70, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Researchers testing OpenAI's textGrain watermark for AI-generated text report that it often fails to survive light editing or texts under about 150 words, raising doubts about whether it can meet the EU AI Act's Article 50 transparency rules, which took effect on 2 August.", "page": "https://rubric.hrih.si/editions/2026-11-07/79/", "card": "https://rubric.hrih.si/og/2026-11-07/79.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.645, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.5883, "spread": 0.145 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.57, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.6833, "spread": 0.21 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.497, "judge_a": 0.5, "judge_b": 0.7, "mean": 0.5656, "spread": 0.203 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.515, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.545, "spread": 0.12 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.627, "judge_a": 0.6, "judge_b": 0.74, "mean": 0.6556, "spread": 0.14 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.735, "judge_a": 0.45, "judge_b": 0.64, "mean": 0.6083, "spread": 0.285 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.8, "judge_b": 0.93, "mean": 0.8733, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0533, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.26, "judge_a": 0.2, "judge_b": 0.2, "mean": 0.22, "spread": 0.06 } } }, { "rank": 80, "id": "st_20261107_17283", "title": "HN: CloneGuard – scan a repo for credential-stealing tricks before you run it", "url": "https://github.com/tharun208/cloneguard", "domain": "github.com", "site": "GitHub", "categories": [ "programming", "security" ], "category_names": [ "Programming & dev tools", "Security" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T19:49:27.000Z", "jev": 0.6782, "judge_a": 0.478, "judge_b": 0.5975, "score": 0.5845, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 71, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "CloneGuard is a dependency-free Go command-line tool that scans a project folder for credential-stealing patterns before anyone runs it. It only reads files and never executes project code.", "page": "https://rubric.hrih.si/editions/2026-11-07/80/", "card": "https://rubric.hrih.si/og/2026-11-07/80.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.405, "judge_a": 0.3, "judge_b": 0.35, "mean": 0.3517, "spread": 0.105 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.72, "judge_a": 0.6, "judge_b": 0.8, "mean": 0.7067, "spread": 0.2 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.73, "judge_a": 0.55, "judge_b": 0.55, "mean": 0.61, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.913, "judge_a": 0.7, "judge_b": 0.75, "mean": 0.7875, "spread": 0.213 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.927, "judge_a": 0.45, "judge_b": 0.55, "mean": 0.6422, "spread": 0.477 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.848, "judge_a": 0.45, "judge_b": 0.58, "mean": 0.6258, "spread": 0.398 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.82, "judge_a": 0.9, "judge_b": 0.95, "mean": 0.89, "spread": 0.13 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.12, "spread": 0.12 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.15, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.1933, "spread": 0.1 } } }, { "rank": 81, "id": "st_20261107_17370", "title": "Teaching AI to build scientific models", "url": "https://www.anthropic.com/research/teaching-ai-to-build-scientific-models", "domain": "anthropic.com", "site": "anthropic.com", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T21:24:28.000Z", "jev": 0.5033, "judge_a": 0.5267, "judge_b": 0.7201, "score": 0.5834, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 72, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A recent essay argues that AI can now do much of scientific modeling, from fitting equations to constraining parameters, but that fit alone is not understanding. It uses Hodgkin and Huxley's 1951 nerve model as the example, then describes simulation-based inference, a method that lets researchers constrain complex simulators with…", "page": "https://rubric.hrih.si/editions/2026-11-07/81/", "card": "https://rubric.hrih.si/og/2026-11-07/81.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.55, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5733, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.5, "judge_a": 0.72, "judge_b": 0.9, "mean": 0.7067, "spread": 0.4 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.637, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.6056, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.41, "judge_a": 0.4, "judge_b": 0.62, "mean": 0.4767, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.433, "judge_a": 0.7, "judge_b": 0.7, "mean": 0.6111, "spread": 0.267 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.745, "judge_a": 0.6, "judge_b": 0.7, "mean": 0.6817, "spread": 0.145 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.9, "judge_b": 0.88, "mean": 0.8933, "spread": 0.02 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.05, "judge_b": 0.03, "mean": 0.0467, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.22, "judge_a": 0.2, "judge_b": 0.18, "mean": 0.2, "spread": 0.04 } } }, { "rank": 82, "id": "st_20261107_16821", "title": "Local tier-1 SOC triage on Wazuh with a fine-tuned Gemma 4", "url": "https://github.com/oleg-vdv/kezekshi-l1", "domain": "github.com", "site": "GitHub", "categories": [ "ai", "security" ], "category_names": [ "AI & machine learning", "Security" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T13:08:35.000Z", "jev": 0.5915, "judge_a": 0.481, "judge_b": 0.6741, "score": 0.5822, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 73, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Kezekshi is an open-source, locally run first-line triage system for Wazuh security alerts, published on Show HN by its author (oleg-vdv). Deterministic rules suppress noise, and a LoRA-fine-tuned Gemma 4 E2B drafts analyst cards for ambiguous alerts, with no data sent externally.", "page": "https://rubric.hrih.si/editions/2026-11-07/82/", "card": "https://rubric.hrih.si/og/2026-11-07/82.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.478, "judge_a": 0.3, "judge_b": 0.55, "mean": 0.4425, "spread": 0.25 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.61, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.6767, "spread": 0.22 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.393, "judge_a": 0.4, "judge_b": 0.34, "mean": 0.3778, "spread": 0.06 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.715, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.6983, "spread": 0.18 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.89, "judge_a": 0.6, "judge_b": 0.7, "mean": 0.73, "spread": 0.29 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.983, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.8475, "spread": 0.283 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.92, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.1, "judge_b": 0.05, "mean": 0.07, "spread": 0.05 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.15, "judge_a": 0.2, "judge_b": 0.04, "mean": 0.13, "spread": 0.16 } } }, { "rank": 83, "id": "st_20261107_16950", "title": "Supercharging XLA-Based Inference: Announcing the Launch of XLA Serving Engine", "url": "https://opensource.googleblog.com/2026/11/supercharging-xla-inference-xla-serving-engine.html", "domain": "opensource.googleblog.com", "site": "Google Open Source Blog", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T15:21:03.000Z", "jev": 0.6025, "judge_a": 0.4337, "judge_b": 0.7057, "score": 0.5806, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 75, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Google has open-sourced XLA Serving Engine (XSE) under Apache 2.0, a C++ inference runtime for the OpenXLA ecosystem that lets teams train models in JAX or PyTorch and serve them without Python.", "page": "https://rubric.hrih.si/editions/2026-11-07/83/", "card": "https://rubric.hrih.si/og/2026-11-07/83.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.67, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.65, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.64, "judge_a": 0.62, "judge_b": 0.86, "mean": 0.7067, "spread": 0.24 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.56, "judge_a": 0.4, "judge_b": 0.63, "mean": 0.53, "spread": 0.23 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.755, "judge_a": 0.55, "judge_b": 0.82, "mean": 0.7083, "spread": 0.27 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.607, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5922, "spread": 0.27 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.74, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.6533, "spread": 0.24 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.93, "judge_b": 0.98, "mean": 0.9367, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.18, "judge_a": 0.4, "judge_b": 0.28, "mean": 0.2867, "spread": 0.22 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.38, "judge_a": 0.35, "judge_b": 0.32, "mean": 0.35, "spread": 0.06 } } }, { "rank": 84, "id": "st_20261107_17397", "title": "The Intensional Model in Disguise", "url": "https://fzakaria.com/2026/11/06/the-intensional-model-in-disguise", "domain": "fzakaria.com", "site": "Farid Zakaria’s Blog", "categories": [ "programming", "devops" ], "category_names": [ "Programming & dev tools", "Infrastructure & DevOps" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T21:57:31.000Z", "jev": 0.5386, "judge_a": 0.5002, "judge_b": 0.6913, "score": 0.5767, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 79, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Fabian Zakaria's post argues that Nix, Bazel and buck2 share a common core: the intensional model, where build outputs are addressed by content rather than by the recipe that produced them.", "page": "https://rubric.hrih.si/editions/2026-11-07/84/", "card": "https://rubric.hrih.si/og/2026-11-07/84.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.52, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.48, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.61, "judge_a": 0.72, "judge_b": 0.78, "mean": 0.7033, "spread": 0.17 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.54, "judge_a": 0.4, "judge_b": 0.58, "mean": 0.5067, "spread": 0.18 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.648, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6392, "spread": 0.17 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.713, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.7444, "spread": 0.12 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.935, "judge_a": 0.8, "judge_b": 0.84, "mean": 0.8583, "spread": 0.135 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.6, "judge_b": 0.93, "mean": 0.8067, "spread": 0.33 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.05, "judge_b": 0.02, "mean": 0.05, "spread": 0.06 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.63, "judge_a": 0.25, "judge_b": 0.18, "mean": 0.3533, "spread": 0.45 } } }, { "rank": 85, "id": "st_20261107_16867", "title": "ToyBlocks: A controlled world for robot learning and evaluation", "url": "https://allenai.org/blog/toyblocks", "domain": "allenai.org", "site": "Ai2 Blog", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T08:00:00.000Z", "jev": 0.5828, "judge_a": 0.4215, "judge_b": 0.725, "score": 0.5764, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 80, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Ai2 has introduced ToyBlocks, the first \"FoundationWorld\": an open, controlled simulation of a robot arm manipulating blocks on a tabletop, built on Isaac Lab. It automatically generates demonstrations and tasks, letting researchers vary training data, test generalization, and trace failures.", "page": "https://rubric.hrih.si/editions/2026-11-07/85/", "card": "https://rubric.hrih.si/og/2026-11-07/85.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.643, "judge_a": 0.4, "judge_b": 0.78, "mean": 0.6075, "spread": 0.38 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.61, "judge_a": 0.6, "judge_b": 0.9, "mean": 0.7033, "spread": 0.3 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.56, "judge_a": 0.35, "judge_b": 0.56, "mean": 0.49, "spread": 0.21 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.473, "judge_a": 0.4, "judge_b": 0.68, "mean": 0.5175, "spread": 0.28 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.94, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.7867, "spread": 0.34 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.835, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.8117, "spread": 0.2 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.89, "judge_a": 0.95, "judge_b": 0.98, "mean": 0.94, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.2, "judge_b": 0.12, "mean": 0.1367, "spread": 0.11 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.21, "judge_a": 0.35, "judge_b": 0.28, "mean": 0.28, "spread": 0.14 } } }, { "rank": 86, "id": "st_20261107_17041", "title": "Promethevs: Full DeepSeek-v4.1-Flash-1017, locally, on 12GB of VRAM", "url": "https://github.com/Magnvs7/Promethevs", "domain": "github.com", "site": "GitHub", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T16:14:47.000Z", "jev": 0.555, "judge_a": 0.4585, "judge_b": 0.7244, "score": 0.5792, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 76, "topic_penalty": 0.08, "repeat_penalty": 0.003, "summary": "PROMETHEVS is a free, open-source engine that runs the full 510 GB DeepSeek-V4.1-Flash-1017 model on a consumer PC with 64 GB RAM and a 12 GB RTX 3060, using a 2-bit copy of experts plus exact correction of words where the model is uncertain.", "page": "https://rubric.hrih.si/editions/2026-11-07/86/", "card": "https://rubric.hrih.si/og/2026-11-07/86.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.593, "judge_a": 0.5, "judge_b": 0.76, "mean": 0.6175, "spread": 0.26 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.62, "judge_a": 0.6, "judge_b": 0.88, "mean": 0.7, "spread": 0.28 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.493, "judge_a": 0.5, "judge_b": 0.68, "mean": 0.5578, "spread": 0.187 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.658, "judge_a": 0.55, "judge_b": 0.83, "mean": 0.6792, "spread": 0.28 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.957, "judge_a": 0.55, "judge_b": 0.9, "mean": 0.8022, "spread": 0.407 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.98, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8867, "spread": 0.18 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.9133, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.35, "judge_b": 0.14, "mean": 0.1967, "spread": 0.25 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.72, "judge_a": 0.45, "judge_b": 0.62, "mean": 0.5967, "spread": 0.27 } } }, { "rank": 87, "id": "st_20261107_17042", "title": "WebAssembly Specification Release 3.0 (2026-11-04)", "url": "https://webassembly.github.io/spec/core/", "domain": "webassembly.github.io", "site": "webassembly.github.io", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T16:13:46.000Z", "jev": 0.5097, "judge_a": 0.5095, "judge_b": 0.7084, "score": 0.5758, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 81, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "The WebAssembly Working Group has published Release 3.0 of the WebAssembly Core Specification, dated 2026-11-04 and edited by Andreas Rossberg. The page is primarily a table of contents linking to the spec's sections on syntax, validation, execution, binary and text formats, and appendices.", "page": "https://rubric.hrih.si/editions/2026-11-07/87/", "card": "https://rubric.hrih.si/og/2026-11-07/87.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.725, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.665, "spread": 0.175 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.37, "judge_a": 0.5, "judge_b": 0.82, "mean": 0.5633, "spread": 0.45 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.667, "judge_a": 0.5, "judge_b": 0.67, "mean": 0.6122, "spread": 0.17 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.523, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.5475, "spread": 0.12 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.677, "judge_a": 0.6, "judge_b": 0.9, "mean": 0.7256, "spread": 0.3 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.235, "judge_a": 0.55, "judge_b": 0.18, "mean": 0.3217, "spread": 0.37 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.84, "judge_a": 0.6, "judge_b": 0.98, "mean": 0.8067, "spread": 0.38 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.17, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.0767, "spread": 0.16 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.05, "judge_b": 0.02, "mean": 0.0767, "spread": 0.14 } } }, { "rank": 88, "id": "st_20261107_16602", "title": "Eski – an in-memory store in Odin, measured against Redis on one core", "url": "https://github.com/vighnesh-bheed/eski", "domain": "github.com", "site": "GitHub", "categories": [ "programming", "data" ], "category_names": [ "Programming & dev tools", "Data" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T08:44:09.000Z", "jev": 0.6, "judge_a": 0.3745, "judge_b": 0.7407, "score": 0.5718, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 82, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Eski is a new in-memory data store written in Odin that speaks the Redis RESP protocol, so existing clients connect unchanged. Its author reports, from a committed benchmark dataset on one server core, higher throughput and no-higher p99 latency than Redis, Valkey and Dragonfly across the tested workloads, with about…", "page": "https://rubric.hrih.si/editions/2026-11-07/88/", "card": "https://rubric.hrih.si/og/2026-11-07/88.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.515, "judge_a": 0.35, "judge_b": 0.62, "mean": 0.495, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.69, "judge_a": 0.6, "judge_b": 0.9, "mean": 0.73, "spread": 0.3 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.453, "judge_a": 0.4, "judge_b": 0.55, "mean": 0.4678, "spread": 0.15 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.685, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.6183, "spread": 0.27 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.993, "judge_a": 0.6, "judge_b": 0.93, "mean": 0.8411, "spread": 0.393 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.7, "judge_b": 0.94, "mean": 0.8792, "spread": 0.298 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.92, "judge_a": 0.8, "judge_b": 0.98, "mean": 0.9, "spread": 0.18 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.08, "judge_a": 0.35, "judge_b": 0.03, "mean": 0.1533, "spread": 0.32 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.39, "judge_a": 0.5, "judge_b": 0.08, "mean": 0.3233, "spread": 0.42 } } }, { "rank": 89, "id": "st_20261107_17049", "title": "The Comprehensive Guide to the Bank of Nvidia", "url": "https://hntrbrk.com/investigations/nvidia", "domain": "hntrbrk.com", "site": "hntrbrk.com", "categories": [ "ai", "hardware", "startups" ], "category_names": [ "AI & machine learning", "Hardware & devices", "Startups & founders" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T16:10:04.000Z", "jev": 0.463, "judge_a": 0.51, "judge_b": 0.7417, "score": 0.5715, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 83, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Hunterbrook Media reports that Nvidia has committed roughly $530.5 billion in future obligations, including supply purchase commitments, neocloud equity injections, data center guarantees and capacity purchase agreements, based on its latest 10-Q filing.", "page": "https://rubric.hrih.si/editions/2026-11-07/89/", "card": "https://rubric.hrih.si/og/2026-11-07/89.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.618, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.6792, "spread": 0.22 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.56, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.72, "spread": 0.34 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.62, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6467, "spread": 0.12 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.275, "judge_a": 0.35, "judge_b": 0.68, "mean": 0.435, "spread": 0.405 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.643, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7078, "spread": 0.137 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.358, "judge_a": 0.2, "judge_b": 0.42, "mean": 0.3258, "spread": 0.22 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.75, "judge_b": 0.94, "mean": 0.8633, "spread": 0.19 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.11, "judge_a": 0.05, "judge_b": 0.01, "mean": 0.0567, "spread": 0.1 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.42, "judge_a": 0.4, "judge_b": 0.34, "mean": 0.3867, "spread": 0.08 } } }, { "rank": 90, "id": "st_20261107_16678", "title": "The Docker group is root access, my agent used it to write to /etc.", "url": "https://jeena.net/notes/1497", "domain": "jeena.net", "site": "jeena.net", "categories": [ "security", "devops", "it" ], "category_names": [ "Security", "Infrastructure & DevOps", "IT & networking" ], "format": "Other", "depth": "three_judges", "published_at": "2026-11-06T10:59:56.000Z", "jev": 0.5424, "judge_a": 0.6105, "judge_b": 0.5562, "score": 0.5697, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 84, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A developer writes that for about ten years they ran Docker with their user in the docker group, which lets any process running as that user mount the host root filesystem and write as root.", "page": "https://rubric.hrih.si/editions/2026-11-07/90/", "card": "https://rubric.hrih.si/og/2026-11-07/90.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.35, "judge_a": 0.4, "judge_b": 0.35, "mean": 0.3667, "spread": 0.05 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.57, "judge_a": 0.82, "judge_b": 0.7, "mean": 0.6967, "spread": 0.25 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.72, "judge_a": 0.6, "judge_b": 0.5, "mean": 0.6067, "spread": 0.22 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.753, "judge_a": 0.8, "judge_b": 0.72, "mean": 0.7575, "spread": 0.08 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.163, "judge_a": 0.35, "judge_b": 0.55, "mean": 0.3544, "spread": 0.387 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.67, "judge_a": 0.45, "judge_b": 0.28, "mean": 0.4667, "spread": 0.39 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.75, "judge_a": 0.7, "judge_b": 0.72, "mean": 0.7233, "spread": 0.05 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.03, "judge_b": 0.01, "mean": 0.03, "spread": 0.04 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.17, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.15, "spread": 0.12 } } }, { "rank": 91, "id": "st_20261107_17196", "title": "Memorilla: Latent Semantic Memory for LLMs", "url": "https://arxiv.org/abs/2610.21375", "domain": "arxiv.org", "site": "arXiv", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T18:03:33.000Z", "jev": 0.6319, "judge_a": 0.611, "judge_b": 0.7646, "score": 0.5691, "contested": false, "flag_penalty": 0, "freshness_penalty": 0.1, "score_rank": 85, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Memorilla is a memory system that gives a frozen LLM query-specific access to an entire document collection, such as user chat histories, book chapters, or an agent's past turns.", "page": "https://rubric.hrih.si/editions/2026-11-07/91/", "card": "https://rubric.hrih.si/og/2026-11-07/91.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.675, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6483, "spread": 0.17 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.63, "judge_a": 0.75, "judge_b": 0.91, "mean": 0.7633, "spread": 0.28 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.663, "judge_a": 0.6, "judge_b": 0.67, "mean": 0.6444, "spread": 0.07 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.643, "judge_a": 0.55, "judge_b": 0.72, "mean": 0.6375, "spread": 0.17 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.967, "judge_a": 0.8, "judge_b": 0.88, "mean": 0.8822, "spread": 0.167 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.97, "judge_a": 0.85, "judge_b": 0.96, "mean": 0.9267, "spread": 0.12 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.9, "judge_b": 0.99, "mean": 0.9333, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.05, "judge_b": 0.04, "mean": 0.0533, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.46, "judge_a": 0.3, "judge_b": 0.22, "mean": 0.3267, "spread": 0.24 } } }, { "rank": 92, "id": "st_20261107_17052", "title": "AI models built outside the US and China are nearly a year behind", "url": "https://village.dev/blog/outside-the-us-and-china-the-world-is-building-last-years-ai/", "domain": "village.dev", "site": "Village", "categories": [ "ai" ], "category_names": [ "AI & machine learning" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T16:10:14.000Z", "jev": 0.523, "judge_a": 0.4255, "judge_b": 0.7538, "score": 0.5673, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 87, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Models from labs outside the US and China have trailed the US frontier by roughly ten months since 2024, with a median gap of 9.9 months, according to a study using Village's LLM Atlas, which scores 857 models on one capability scale from public benchmarks.", "page": "https://rubric.hrih.si/editions/2026-11-07/92/", "card": "https://rubric.hrih.si/og/2026-11-07/92.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.583, "judge_a": 0.45, "judge_b": 0.75, "mean": 0.5942, "spread": 0.3 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.6, "judge_a": 0.6, "judge_b": 0.9, "mean": 0.7, "spread": 0.3 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.697, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6589, "spread": 0.28 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.368, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.4292, "spread": 0.32 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.817, "judge_a": 0.55, "judge_b": 0.85, "mean": 0.7389, "spread": 0.3 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.608, "judge_a": 0.3, "judge_b": 0.72, "mean": 0.5425, "spread": 0.42 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.8, "judge_b": 0.98, "mean": 0.88, "spread": 0.18 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.1, "judge_b": 0.03, "mean": 0.0633, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.43, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.3033, "spread": 0.25 } } }, { "rank": 93, "id": "st_20261107_17467", "title": "Outbound SIP calling is now available to all organizations", "url": "https://developers.openai.com/api/docs/guides/voice-sip", "domain": "developers.openai.com", "site": "OpenAI Developers", "categories": [ "ai", "it" ], "category_names": [ "AI & machine learning", "IT & networking" ], "format": "Tutorial", "depth": "claim_checked", "published_at": "2026-11-06T23:03:55.000Z", "jev": 0.6528, "judge_a": 0.5635, "judge_b": 0.7838, "score": 0.5666, "contested": false, "flag_penalty": 0.1, "freshness_penalty": 0, "score_rank": 88, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "OpenAI has made outbound SIP calling available to all organizations through the Live API, letting developers place phone calls through a SIP trunk provider such as Twilio while GPT-Live handles the conversation.", "page": "https://rubric.hrih.si/editions/2026-11-07/93/", "card": "https://rubric.hrih.si/og/2026-11-07/93.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.573, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5808, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.57, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.6767, "spread": 0.29 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.463, "judge_a": 0.5, "judge_b": 0.55, "mean": 0.5044, "spread": 0.087 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.928, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8758, "spread": 0.127 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.833, "judge_a": 0.7, "judge_b": 0.95, "mean": 0.8278, "spread": 0.25 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.995, "judge_a": 0.85, "judge_b": 0.82, "mean": 0.8883, "spread": 0.175 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.91, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.93, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.25, "judge_b": 0.08, "mean": 0.1333, "spread": 0.18 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.19, "judge_a": 0.15, "judge_b": 0.04, "mean": 0.1267, "spread": 0.15 } } }, { "rank": 94, "id": "st_20261107_17202", "title": "PyPI's redesigned project pages with clearer security signals are now live", "url": "https://blog.pypi.org/posts/2026-11-06-ui-updates-complete/", "domain": "blog.pypi.org", "site": "PyPI Blog", "categories": [ "programming", "security", "design" ], "category_names": [ "Programming & dev tools", "Security", "Design & UX" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T06:09:00.000Z", "jev": 0.6354, "judge_a": 0.4965, "judge_b": 0.5661, "score": 0.566, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 89, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "PyPI has launched redesigned project pages that make package security signals more visible. The update adds a dedicated security tab showing per-file attestations and provenance states, which indicate whether release files can be cryptographically traced to their source.", "page": "https://rubric.hrih.si/editions/2026-11-07/94/", "card": "https://rubric.hrih.si/og/2026-11-07/94.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.655, "judge_a": 0.4, "judge_b": 0.55, "mean": 0.535, "spread": 0.255 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.55, "judge_a": 0.55, "judge_b": 0.58, "mean": 0.56, "spread": 0.03 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.623, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.6144, "spread": 0.023 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.783, "judge_a": 0.6, "judge_b": 0.58, "mean": 0.6542, "spread": 0.203 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.75, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.7233, "spread": 0.22 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.73, "judge_a": 0.45, "judge_b": 0.43, "mean": 0.5367, "spread": 0.3 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.8, "judge_b": 0.96, "mean": 0.8867, "spread": 0.16 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.15, "judge_b": 0.14, "mean": 0.1367, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.13, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.12, "spread": 0.07 } } }, { "rank": 95, "id": "st_20261107_16497", "title": "TieredQ: Space Efficient Vector Quantization (Beating RaBitQ and TurboQuant)", "url": "https://github.com/SeanPedersen/TieredQ", "domain": "github.com", "site": "GitHub", "categories": [ "ai", "data" ], "category_names": [ "AI & machine learning", "Data" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T05:22:43.000Z", "jev": 0.597, "judge_a": 0.372, "judge_b": 0.725, "score": 0.5647, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 92, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "TieredQ is a vector quantization algorithm that applies PCA, then greedily assigns between 0 and 8 (up to 12 with fine tiers) bits per principal dimension according to variance, followed by per-dimension scalar quantization.", "page": "https://rubric.hrih.si/editions/2026-11-07/95/", "card": "https://rubric.hrih.si/og/2026-11-07/95.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.565, "judge_a": 0.35, "judge_b": 0.68, "mean": 0.5317, "spread": 0.33 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.61, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.6567, "spread": 0.36 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.47, "judge_a": 0.3, "judge_b": 0.52, "mean": 0.43, "spread": 0.22 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.825, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.7483, "spread": 0.225 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.8, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6933, "spread": 0.3 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.995, "judge_a": 0.8, "judge_b": 0.98, "mean": 0.925, "spread": 0.195 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.86, "judge_a": 0.9, "judge_b": 0.94, "mean": 0.9, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.11, "judge_a": 0.3, "judge_b": 0.12, "mean": 0.1767, "spread": 0.19 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.47, "judge_a": 0.6, "judge_b": 0.2, "mean": 0.4233, "spread": 0.4 } } }, { "rank": 96, "id": "st_20261107_16648", "title": "Technical/eng breakdown of Framer's Lighthouse 13.6 optimizations", "url": "https://www.framer.com/blog/lighthouse-13-6/", "domain": "framer.com", "site": "Framer", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "Write-up", "depth": "three_judges", "published_at": "2026-11-06T09:35:41.000Z", "jev": 0.6008, "judge_a": 0.3565, "judge_b": 0.7356, "score": 0.5643, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 93, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Framer's engineering team, working with Google, describes two fixes shipped in Lighthouse 13.6 that stop its slow-phone simulation from penalizing preloaded resources and text-based LCP. Preloads and fonts no longer delay First Contentful Paint, and text LCP is set equal to FCP when they occur in the same paint.", "page": "https://rubric.hrih.si/editions/2026-11-07/96/", "card": "https://rubric.hrih.si/og/2026-11-07/96.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.6, "judge_a": 0.35, "judge_b": 0.67, "mean": 0.54, "spread": 0.32 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.5, "judge_b": 0.86, "mean": 0.67, "spread": 0.36 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.58, "judge_a": 0.4, "judge_b": 0.72, "mean": 0.5667, "spread": 0.32 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.698, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6592, "spread": 0.28 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.787, "judge_a": 0.7, "judge_b": 0.86, "mean": 0.7822, "spread": 0.16 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.888, "judge_a": 0.7, "judge_b": 0.82, "mean": 0.8025, "spread": 0.188 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.92, "judge_a": 0.8, "judge_b": 0.98, "mean": 0.9, "spread": 0.18 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.13, "judge_a": 0.45, "judge_b": 0.12, "mean": 0.2333, "spread": 0.33 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.42, "judge_a": 0.4, "judge_b": 0.22, "mean": 0.3467, "spread": 0.2 } } }, { "rank": 97, "id": "st_20261107_16397", "title": "PCAP-LM: Helping AI assistants read network traffic", "url": "https://blog.apnic.net/2026/11/06/pcap-lm-helping-ai-assistants-read-network-traffic/", "domain": "blog.apnic.net", "site": "APNIC Blog", "categories": [ "ai", "it" ], "category_names": [ "AI & machine learning", "IT & networking" ], "format": "Paper", "depth": "three_judges", "published_at": "2026-11-06T00:04:21.000Z", "jev": 0.5591, "judge_a": 0.428, "judge_b": 0.7036, "score": 0.5635, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 94, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "PCAP-LM, described by Xavier Marjou of Orange, rewrites network packet captures as compact symbolic text so AI assistants can read an entire capture in one pass. Using a \"PacketGlyphs\" notation with run-length encoding and token merging, typical captures shrink roughly 800 times versus Wireshark's verbose text output.", "page": "https://rubric.hrih.si/editions/2026-11-07/97/", "card": "https://rubric.hrih.si/og/2026-11-07/97.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.523, "judge_a": 0.35, "judge_b": 0.66, "mean": 0.5108, "spread": 0.31 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.6, "judge_b": 0.88, "mean": 0.71, "spread": 0.28 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.573, "judge_a": 0.4, "judge_b": 0.48, "mean": 0.4844, "spread": 0.173 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.58, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.6, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.79, "judge_a": 0.55, "judge_b": 0.78, "mean": 0.7067, "spread": 0.24 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.815, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.7117, "spread": 0.215 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9167, "spread": 0.11 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.12, "judge_a": 0.2, "judge_b": 0.08, "mean": 0.1333, "spread": 0.12 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.35, "judge_a": 0.35, "judge_b": 0.12, "mean": 0.2733, "spread": 0.23 } } }, { "rank": 98, "id": "st_20261107_16811", "title": "Baton – Keep long coding goals moving across AI coding agents", "url": "https://github.com/cahian/baton", "domain": "github.com", "site": "GitHub", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T13:24:08.000Z", "jev": 0.6769, "judge_a": 0.4655, "judge_b": 0.5416, "score": 0.5613, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 95, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Baton is an open-source tool that keeps long coding goals moving across Codex, Antigravity and Claude API. It runs one vendor's own CLI until a configured usage or spend limit is hit, then hands the task, along with a sanitized transcript and session IDs, to the next vendor.", "page": "https://rubric.hrih.si/editions/2026-11-07/98/", "card": "https://rubric.hrih.si/og/2026-11-07/98.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.543, "judge_a": 0.35, "judge_b": 0.42, "mean": 0.4375, "spread": 0.193 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.6, "judge_b": 0.78, "mean": 0.7033, "spread": 0.18 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.513, "judge_a": 0.4, "judge_b": 0.42, "mean": 0.4444, "spread": 0.113 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.863, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.7275, "spread": 0.263 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.747, "judge_a": 0.55, "judge_b": 0.58, "mean": 0.6256, "spread": 0.197 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.8, "judge_b": 0.78, "mean": 0.8592, "spread": 0.218 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.9, "judge_a": 0.9, "judge_b": 0.98, "mean": 0.9267, "spread": 0.08 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.2, "judge_b": 0.38, "mean": 0.2167, "spread": 0.31 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.17, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.2167, "spread": 0.13 } } }, { "rank": 99, "id": "st_20261107_17180", "title": "We Know 15,579 Websites Are Infected. How Do We Tell Them?", "url": "https://scotthelme.co.uk/we-know-15-579-websites-are-infected-how-do-we-tell-them/", "domain": "scotthelme.co.uk", "site": "Scott Helme", "categories": [ "security" ], "category_names": [ "Security" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T18:13:35.000Z", "jev": 0.5498, "judge_a": 0.507, "judge_b": 0.6231, "score": 0.5599, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 96, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Report URI's Scott Helme says his threat intelligence team has identified 15,579 websites infected with malicious JavaScript, collectively receiving an estimated 135 million monthly visits. Discovery now outpaces responsible disclosure, since most sites lack contact details and rarely respond. He is asking for help turning the data into action.", "page": "https://rubric.hrih.si/editions/2026-11-07/99/", "card": "https://rubric.hrih.si/og/2026-11-07/99.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.598, "judge_a": 0.5, "judge_b": 0.62, "mean": 0.5725, "spread": 0.12 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.6, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.6933, "spread": 0.18 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.633, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.6178, "spread": 0.033 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.498, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.5925, "spread": 0.183 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.447, "judge_a": 0.5, "judge_b": 0.58, "mean": 0.5089, "spread": 0.133 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.573, "judge_a": 0.4, "judge_b": 0.43, "mean": 0.4675, "spread": 0.173 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.8, "judge_b": 0.94, "mean": 0.8733, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.3, "judge_b": 0.18, "mean": 0.1933, "spread": 0.2 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.3, "judge_b": 0.2, "mean": 0.22, "spread": 0.14 } } }, { "rank": 100, "id": "st_20261107_16598", "title": "How AI Is Changing Software Development in 2026", "url": "https://palettevault.github.io/blog/ai-development-2026/", "domain": "palettevault.github.io", "site": "Palette Vault", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T08:59:24.000Z", "jev": 0.7583, "judge_a": 0.295, "judge_b": 0.6261, "score": 0.5598, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 97, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "A feature-length piece on how AI coding agents are changing software development in 2026 describes a shift from chat-based code snippets to agents that inspect repositories, edit multiple files, run tests, and iterate on tasks.", "page": "https://rubric.hrih.si/editions/2026-11-07/100/", "card": "https://rubric.hrih.si/og/2026-11-07/100.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.743, "judge_a": 0.3, "judge_b": 0.48, "mean": 0.5075, "spread": 0.443 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.7, "judge_a": 0.35, "judge_b": 0.68, "mean": 0.5767, "spread": 0.35 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.943, "judge_a": 0.6, "judge_b": 0.9, "mean": 0.8144, "spread": 0.343 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.91, "judge_a": 0.5, "judge_b": 0.72, "mean": 0.71, "spread": 0.41 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.593, "judge_a": 0.45, "judge_b": 0.57, "mean": 0.5378, "spread": 0.143 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.703, "judge_a": 0.3, "judge_b": 0.38, "mean": 0.4608, "spread": 0.403 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.2, "judge_b": 0.64, "mean": 0.5733, "spread": 0.68 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.4, "judge_b": 0.04, "mean": 0.1767, "spread": 0.36 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.15, "judge_a": 0.25, "judge_b": 0.12, "mean": 0.1733, "spread": 0.13 } } }, { "rank": 101, "id": "st_20261107_16660", "title": "Google Donates Agent Substrate and GVisor to the CNCF", "url": "https://opensource.googleblog.com/2026/11/sandboxing-untrusted-ai-agents-at-scale.html", "domain": "opensource.googleblog.com", "site": "Google Open Source Blog", "categories": [ "programming", "ai", "devops" ], "category_names": [ "Programming & dev tools", "AI & machine learning", "Infrastructure & DevOps" ], "format": "Release", "depth": "three_judges", "published_at": "2026-11-06T10:19:21.000Z", "jev": 0.6684, "judge_a": 0.5616, "judge_b": 0.5498, "score": 0.5932, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 65, "topic_penalty": 0.08, "repeat_penalty": 0.0344, "summary": "Google is donating Agent Substrate, an orchestrator for AI agent workloads, and the gVisor application kernel to the Cloud Native Computing Foundation at KubeCon North America, alongside the OpenRL training project.", "page": "https://rubric.hrih.si/editions/2026-11-07/101/", "card": "https://rubric.hrih.si/og/2026-11-07/101.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.728, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6825, "spread": 0.128 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.76, "judge_a": 0.8, "judge_b": 0.84, "mean": 0.8, "spread": 0.08 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.683, "judge_a": 0.67, "judge_b": 0.68, "mean": 0.6778, "spread": 0.013 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.748, "judge_a": 0.7, "judge_b": 0.76, "mean": 0.7358, "spread": 0.06 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.467, "judge_a": 0.4, "judge_b": 0.62, "mean": 0.4956, "spread": 0.22 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.773, "judge_a": 0.6, "judge_b": 0.72, "mean": 0.6975, "spread": 0.173 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.85, "judge_b": 0.96, "mean": 0.8967, "spread": 0.11 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.2, "judge_a": 0.45, "judge_b": 0.78, "mean": 0.4767, "spread": 0.58 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.23, "judge_a": 0.35, "judge_b": 0.58, "mean": 0.3867, "spread": 0.35 } } }, { "rank": 102, "id": "st_20261107_17423", "title": "Speed Up Image Generation with FLUX 3 Image Klein on NVIDIA Blackwell GPUs", "url": "https://developer.nvidia.com/blog/speed-up-image-generation-with-flux-3-image-klein-on-nvidia-blackwell-gpus/", "domain": "developer.nvidia.com", "site": "NVIDIA Technical Blog", "categories": [ "ai", "hardware" ], "category_names": [ "AI & machine learning", "Hardware & devices" ], "format": "Tutorial", "depth": "three_judges", "published_at": "2026-11-06T22:20:20.000Z", "jev": 0.6556, "judge_a": 0.3365, "judge_b": 0.6798, "score": 0.5573, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 98, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "NVIDIA's TensorRT LLM VisualGen tutorial describes how to speed up FLUX 3 Image Klein, a Black Forest Labs image generation and editing model, on NVIDIA Blackwell GPUs. It covers NVFP4 quantization, optimized GEMM kernels, quantized and sparse attention, step caching, and multi-GPU execution.", "page": "https://rubric.hrih.si/editions/2026-11-07/102/", "card": "https://rubric.hrih.si/og/2026-11-07/102.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.585, "judge_a": 0.3, "judge_b": 0.72, "mean": 0.535, "spread": 0.42 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.65, "judge_a": 0.4, "judge_b": 0.82, "mean": 0.6233, "spread": 0.42 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.43, "judge_a": 0.35, "judge_b": 0.42, "mean": 0.4, "spread": 0.08 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.88, "judge_a": 0.55, "judge_b": 0.82, "mean": 0.75, "spread": 0.33 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.913, "judge_a": 0.6, "judge_b": 0.86, "mean": 0.7911, "spread": 0.313 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.998, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8992, "spread": 0.198 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.88, "judge_a": 0.85, "judge_b": 0.92, "mean": 0.8833, "spread": 0.07 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.11, "judge_a": 0.45, "judge_b": 0.28, "mean": 0.28, "spread": 0.34 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.3, "judge_b": 0.2, "mean": 0.2333, "spread": 0.1 } } }, { "rank": 103, "id": "st_20261107_16540", "title": "Mantis – AI agents browse 36% faster than with Playwright MCP", "url": "https://github.com/perzianhq/mantis", "domain": "github.com", "site": "GitHub", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T06:38:36.000Z", "jev": 0.6063, "judge_a": 0.424, "judge_b": 0.6333, "score": 0.5545, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 101, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Perzian has released Mantis, an open-source MCP server (MIT-licensed, v0.1.x beta) that lets AI agents browse the web by reading short, numbered text instead of screenshots or the accessibility tree.", "page": "https://rubric.hrih.si/editions/2026-11-07/103/", "card": "https://rubric.hrih.si/og/2026-11-07/103.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.525, "judge_a": 0.3, "judge_b": 0.45, "mean": 0.425, "spread": 0.225 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.73, "judge_a": 0.6, "judge_b": 0.85, "mean": 0.7267, "spread": 0.25 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.51, "judge_a": 0.45, "judge_b": 0.55, "mean": 0.5033, "spread": 0.1 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.823, "judge_a": 0.7, "judge_b": 0.78, "mean": 0.7675, "spread": 0.123 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.967, "judge_a": 0.6, "judge_b": 0.8, "mean": 0.7889, "spread": 0.367 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.775, "judge_a": 0.6, "judge_b": 0.62, "mean": 0.665, "spread": 0.175 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.82, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.8933, "spread": 0.14 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.1, "judge_a": 0.35, "judge_b": 0.2, "mean": 0.2167, "spread": 0.25 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.6, "judge_a": 0.5, "judge_b": 0.18, "mean": 0.4267, "spread": 0.42 } } }, { "rank": 104, "id": "st_20261107_17388", "title": "Leeroy: Making AI agents play World of Warcraft", "url": "https://pthm.dev/leeroy-making-ai-agents-play-world-of-warcraft/", "domain": "pthm.dev", "site": "pthm.dev", "categories": [ "programming", "ai" ], "category_names": [ "Programming & dev tools", "AI & machine learning" ], "format": "New project", "depth": "three_judges", "published_at": "2026-11-06T21:07:44.000Z", "jev": 0.5454, "judge_a": 0.438, "judge_b": 0.6778, "score": 0.5537, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 103, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Leeroy is a benchmark by developer Pthm that tests AI coding agents by having them play World of Warcraft through a headless client and Lua scripts. Inspired by RuneBench, it measures how agents plan, recover from mistakes, and coordinate on group content, using XP per hour as the main metric.…", "page": "https://rubric.hrih.si/editions/2026-11-07/104/", "card": "https://rubric.hrih.si/og/2026-11-07/104.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.478, "judge_a": 0.3, "judge_b": 0.58, "mean": 0.4525, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.61, "judge_a": 0.7, "judge_b": 0.9, "mean": 0.7367, "spread": 0.29 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.53, "judge_a": 0.4, "judge_b": 0.42, "mean": 0.45, "spread": 0.13 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.49, "judge_a": 0.4, "judge_b": 0.62, "mean": 0.5033, "spread": 0.22 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.89, "judge_a": 0.5, "judge_b": 0.82, "mean": 0.7367, "spread": 0.39 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.765, "judge_a": 0.6, "judge_b": 0.82, "mean": 0.7283, "spread": 0.22 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.9, "judge_b": 0.96, "mean": 0.91, "spread": 0.09 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.07, "judge_a": 0.15, "judge_b": 0.04, "mean": 0.0867, "spread": 0.11 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.16, "judge_a": 0.25, "judge_b": 0.08, "mean": 0.1633, "spread": 0.17 } } }, { "rank": 105, "id": "st_20261107_17277", "title": "Nondeterminism in Futhark", "url": "https://futhark-lang.org/blog/2026-11-04-nondeterminism.html", "domain": "futhark-lang.org", "site": "futhark-lang.org", "categories": [ "programming" ], "category_names": [ "Programming & dev tools" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T19:16:01.000Z", "jev": 0.4826, "judge_a": 0.5018, "judge_b": 0.6725, "score": 0.5523, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 105, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "Futhark's reduce and similar parallel combinators are documented as producing an arbitrary but unspecified order of applying a nonassociative operator, which means floating-point sums can differ across backends and hardware.", "page": "https://rubric.hrih.si/editions/2026-11-07/105/", "card": "https://rubric.hrih.si/og/2026-11-07/105.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.483, "judge_a": 0.3, "judge_b": 0.58, "mean": 0.4542, "spread": 0.28 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.45, "judge_a": 0.62, "judge_b": 0.8, "mean": 0.6233, "spread": 0.35 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.393, "judge_a": 0.35, "judge_b": 0.38, "mean": 0.3744, "spread": 0.043 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.525, "judge_a": 0.55, "judge_b": 0.7, "mean": 0.5917, "spread": 0.175 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.517, "judge_a": 0.72, "judge_b": 0.82, "mean": 0.6856, "spread": 0.303 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.905, "judge_a": 0.8, "judge_b": 0.9, "mean": 0.8683, "spread": 0.105 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.87, "judge_a": 0.9, "judge_b": 0.97, "mean": 0.9133, "spread": 0.1 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.05, "judge_a": 0.03, "judge_b": 0.02, "mean": 0.0333, "spread": 0.03 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.2, "judge_a": 0.12, "judge_b": 0.08, "mean": 0.1333, "spread": 0.12 } } }, { "rank": 106, "id": "st_20261107_17061", "title": "Reading the Anthropic S-1, with Zvi Mowshowitz", "url": "https://www.complexsystemspodcast.com/episodes/reading-the-anthropic-s-1-with-zvi-mowshowitz/", "domain": "complexsystemspodcast.com", "site": "Complex Systems", "categories": [ "ai", "startups" ], "category_names": [ "AI & machine learning", "Startups & founders" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T16:04:58.000Z", "jev": 0.565, "judge_a": 0.498, "judge_b": 0.5915, "score": 0.5515, "contested": true, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 106, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "The Complex Systems episode with Zvi Mowshowitz reads Anthropic's S-1 filing, focusing on its founder-controlled share structure (a single Class F share holding 50.1% of votes), an uncapped 3:1 employee donation match, and roughly $800–900 billion in forward compute commitments.", "page": "https://rubric.hrih.si/editions/2026-11-07/106/", "card": "https://rubric.hrih.si/og/2026-11-07/106.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.56, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5767, "spread": 0.27 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.79, "judge_a": 0.85, "judge_b": 0.91, "mean": 0.85, "spread": 0.12 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.837, "judge_a": 0.7, "judge_b": 0.68, "mean": 0.7389, "spread": 0.157 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.405, "judge_a": 0.35, "judge_b": 0.58, "mean": 0.445, "spread": 0.23 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.567, "judge_a": 0.45, "judge_b": 0.62, "mean": 0.5456, "spread": 0.17 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.535, "judge_a": 0.2, "judge_b": 0.32, "mean": 0.3517, "spread": 0.335 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.71, "judge_a": 0.6, "judge_b": 0.05, "mean": 0.4533, "spread": 0.66 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.09, "judge_a": 0.15, "judge_b": 0.08, "mean": 0.1067, "spread": 0.07 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.42, "judge_a": 0.3, "judge_b": 0.62, "mean": 0.4467, "spread": 0.32 } } }, { "rank": 107, "id": "st_20261107_17396", "title": "Agent Harness: Definition", "url": "https://www.nngroup.com/articles/agent-harness/", "domain": "nngroup.com", "site": "Nielsen Norman Group", "categories": [ "ai", "product", "design" ], "category_names": [ "AI & machine learning", "Product", "Design & UX" ], "format": "Analysis", "depth": "three_judges", "published_at": "2026-11-06T18:00:00.000Z", "jev": 0.6198, "judge_a": 0.4285, "judge_b": 0.6033, "score": 0.5505, "contested": false, "flag_penalty": 0, "freshness_penalty": 0, "score_rank": 107, "topic_penalty": 0.08, "repeat_penalty": 0, "summary": "An agent harness is the software layer around an AI model that runs the loop of requesting tools, checking permissions, returning results, and enforcing limits.", "page": "https://rubric.hrih.si/editions/2026-11-07/107/", "card": "https://rubric.hrih.si/og/2026-11-07/107.png", "breakdown": { "significance": { "name": "Significance", "weight": 0.24, "jev": 0.63, "judge_a": 0.3, "judge_b": 0.48, "mean": 0.47, "spread": 0.33 }, "reader_interest": { "name": "Reader interest", "weight": 0.26, "jev": 0.5, "judge_a": 0.45, "judge_b": 0.72, "mean": 0.5567, "spread": 0.27 }, "audience_breadth": { "name": "Audience breadth", "weight": 0.13, "jev": 0.71, "judge_a": 0.5, "judge_b": 0.78, "mean": 0.6633, "spread": 0.28 }, "practical_relevance": { "name": "Practical relevance", "weight": 0.22, "jev": 0.843, "judge_a": 0.6, "judge_b": 0.68, "mean": 0.7075, "spread": 0.243 }, "substantiated": { "name": "Substantiated", "weight": 0.05, "jev": 0.157, "judge_a": 0.35, "judge_b": 0.38, "mean": 0.2956, "spread": 0.223 }, "technical_depth": { "name": "Technical depth", "weight": 0.05, "jev": 0.713, "judge_a": 0.5, "judge_b": 0.52, "mean": 0.5775, "spread": 0.213 }, "new_information": { "name": "New information", "weight": 0.05, "jev": 0.75, "judge_a": 0.6, "judge_b": 0.35, "mean": 0.5667, "spread": 0.4 }, "promotional": { "name": "Promotional", "weight": -0.19, "jev": 0.06, "judge_a": 0.1, "judge_b": 0.02, "mean": 0.06, "spread": 0.08 }, "overclaim": { "name": "Overclaim", "weight": -0.11, "jev": 0.08, "judge_a": 0.1, "judge_b": 0.08, "mean": 0.0867, "spread": 0.02 } } } ] }