Detection Is Not Enough: Reasoning Traces as Testable Program Hypotheses
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
A paper on reference-free verification of LLM reasoning traces finds that checking a trace against itself tops out at AUROC 0.741, because most wrong traces contain no false equation.
Score
0.65total under rubric v2.0
- Significance0.58
- Reader interest0.68
- Audience breadth0.49
- Practical relevance0.60
- Substantiated0.85
- Technical depth0.91
- New information0.86
- Counts against: Promotional0.03
- Counts against: Overclaim0.19
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) |
|---|---|---|---|---|---|---|---|
| Significance | 0.59 | 0.42 | 0.72 | 0.58 | 0.30 | +0.25 | +0.144 |
| Reader interest | 0.55 | 0.62 | 0.86 | 0.68 | 0.31 | +0.20 | +0.135 |
| Audience breadth | 0.53 | 0.45 | 0.48 | 0.49 | 0.08 | +0.10 | +0.049 |
| Practical relevance | 0.62 | 0.55 | 0.62 | 0.60 | 0.07 | +0.15 | +0.090 |
| Substantiated | 0.98 | 0.72 | 0.86 | 0.85 | 0.26 | +0.10 | +0.085 |
| Technical depth | 0.99 | 0.80 | 0.94 | 0.91 | 0.19 | +0.10 | +0.091 |
| New information | 0.91 | 0.70 | 0.96 | 0.86 | 0.26 | +0.10 | +0.086 |
| Promotional | 0.03 | 0.03 | 0.02 | 0.03 | 0.01 | −0.15 | −0.004 |
| Overclaim | 0.23 | 0.20 | 0.14 | 0.19 | 0.09 | −0.15 | −0.028 |
| Sum of contributions | +0.647 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.647 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN —
Recorded for context only, checked 24 Oct 2026, 08:54 UTC. It played no part in the ranking.