StateBench: Testing how language models track state changes over long documents
Comments on “StateBench: Testing how language models track state changes over long documents”
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
StateBench, released by Allen Institute for AI researchers, is a benchmark that tests how well language models track evolving entity states across long documents, using four difficulty levels from immutable assignment to general finite swaps.
Why it ranked
No. 73 of 137 with a score of 0.61. The panel rated it highest on new information (0.90) and technical depth (0.82), lowest on practical relevance (0.54). Placed No. 73, No. 65 by score alone: the 24th AI & machine learning story of the day (−0.080). The placement terms move stories; they are not part of the score.
- AI judgeJudge 2claude-haiku-5-5
- Technical AI research on state tracking in LLM architectures with detailed methodology and benchmarks; substantive and novel, but niche to ML researchers, with some unverifiable model names suggesting possible fabrication and modest practical relevance for most builders.
- AI judgeJudge 3gpt-5.6-luna
- A technically detailed benchmark and empirical study of long-context state tracking, comparing transformer, recurrent, and hybrid architectures. It offers novel evidence and a released evaluation framework, with broad implications for model design, though some conclusions remain explicitly open.
Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.
Score
0.61total under rubric v2.3
- Significance0.61
- Reader interest0.72
- Audience breadth0.57
- Practical relevance0.54
- Substantiated0.78
- Technical depth0.82
- New information0.90
- Counts against: Promotional0.08
- Counts against: Overclaim0.24
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| SignificanceHow much does this change what people in tech know or can do? | 0.64 | 0.40 | 0.78 | 0.61 | 0.38 | +0.24 | +0.145 |
| Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on? | 0.65 | 0.60 | 0.90 | 0.72 | 0.30 | +0.26 | +0.186 |
| Audience breadthHow many people who build technology does it concern? | 0.63 | 0.35 | 0.72 | 0.57 | 0.37 | +0.13 | +0.074 |
| Practical relevanceHow useful is it to someone who builds software or tech products? | 0.54 | 0.35 | 0.72 | 0.54 | 0.37 | +0.22 | +0.118 |
| SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations. | 0.85 | 0.60 | 0.90 | 0.78 | 0.30 | +0.05 | +0.039 |
| Technical depthHow technically substantive is it? | 0.80 | 0.75 | 0.92 | 0.82 | 0.17 | +0.05 | +0.041 |
| New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)? | 0.90 | 0.85 | 0.96 | 0.90 | 0.11 | +0.05 | +0.045 |
| PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional. | 0.07 | 0.10 | 0.08 | 0.08 | 0.03 | −0.19 | −0.016 |
| OverclaimDo the headline or conclusions claim more than the body supports? | 0.25 | 0.30 | 0.18 | 0.24 | 0.12 | −0.11 | −0.027 |
| Sum of contributions | +0.607 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.607 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-11-04/73/"><img src="https://rubric.hrih.si/badge/2026-11-04/73.svg" alt="Ranked #73 on Rubric, 4 Nov 2026" width="238" height="28"></a>Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 2 pts, not on front page
Recorded for context only, checked 4 Nov 2026, 05:47 UTC. It played no part in the ranking.