Direct Pass@$k$ Optimization for Thought Pruning: Balancing Quality and Diversity in Test-Time Reasoning
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
PKTP is a test-time pruning method for reasoning models that, after generating N candidate rollouts up to a token cutoff, keeps the subset of k rollouts that maximizes pass@k.
Score
0.65total under rubric v2.0
- Significance0.59
- Reader interest0.64
- Audience breadth0.51
- Practical relevance0.61
- Substantiated0.85
- Technical depth0.94
- New information0.90
- Counts against: Promotional0.03
- Counts against: Overclaim0.20
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 | 0.60 | 0.45 | 0.72 | 0.59 | 0.27 | +0.25 | +0.148 |
| Reader interest | 0.56 | 0.50 | 0.86 | 0.64 | 0.36 | +0.20 | +0.128 |
| Audience breadth | 0.56 | 0.40 | 0.58 | 0.51 | 0.18 | +0.10 | +0.051 |
| Practical relevance | 0.67 | 0.45 | 0.70 | 0.61 | 0.25 | +0.15 | +0.091 |
| Substantiated | 0.95 | 0.75 | 0.84 | 0.85 | 0.20 | +0.10 | +0.085 |
| Technical depth | 0.99 | 0.85 | 0.97 | 0.94 | 0.14 | +0.10 | +0.094 |
| New information | 0.91 | 0.80 | 0.98 | 0.90 | 0.18 | +0.10 | +0.090 |
| Promotional | 0.05 | 0.03 | 0.01 | 0.03 | 0.04 | −0.15 | −0.004 |
| Overclaim | 0.23 | 0.20 | 0.18 | 0.20 | 0.05 | −0.15 | −0.030 |
| Sum of contributions | +0.651 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.651 | ||||||
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.