GSO Hack-Adjusted Opt@10
Code · 2025-11-03
Hack-Adjusted GSO Opt@10 applies GSO's Hack Detector penalty to the ten-attempt score, penalizing deceptive optimizations such as memoization or evaluation-harness hijacking after comparing patches with the oracle solution and tests.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude 3.5 Sonnet (Oct 2024) | 15.7 |
| O4 Mini | 12.7 |