Feedback That Backfires in Games: Do LLM Players Repeat the Move They Just Watched Fail?
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
In games like Sokoban, Connect Four and Othello, small instruction-tuned LLMs were shown a record of their own failed move plus the referee's message, then asked to move again.
Score
0.66total under rubric v2.0
- Significance0.59
- Reader interest0.68
- Audience breadth0.56
- Practical relevance0.69
- Substantiated0.85
- Technical depth0.92
- New information0.85
- Counts against: Promotional0.03
- Counts against: Overclaim0.27
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 | 0.45 | 0.72 | 0.59 | 0.27 | +0.25 | +0.146 |
| Reader interest | 0.54 | 0.60 | 0.90 | 0.68 | 0.36 | +0.20 | +0.136 |
| Audience breadth | 0.56 | 0.50 | 0.62 | 0.56 | 0.12 | +0.10 | +0.056 |
| Practical relevance | 0.70 | 0.60 | 0.78 | 0.69 | 0.18 | +0.15 | +0.104 |
| Substantiated | 0.99 | 0.70 | 0.86 | 0.85 | 0.29 | +0.10 | +0.085 |
| Technical depth | 1.00 | 0.80 | 0.97 | 0.92 | 0.20 | +0.10 | +0.092 |
| New information | 0.88 | 0.70 | 0.98 | 0.85 | 0.28 | +0.10 | +0.085 |
| Promotional | 0.04 | 0.05 | 0.01 | 0.03 | 0.04 | −0.15 | −0.005 |
| Overclaim | 0.36 | 0.25 | 0.20 | 0.27 | 0.16 | −0.15 | −0.041 |
| Sum of contributions | +0.659 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.659 | ||||||
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.