CL4SE - Patch Correctness Assessment (2-Shot): leaderboard
Metric: Accuracy (%) of classifying 2,274 plausible Defects4J patches (1,105 correct, 1,169 overfitting) as correct or overfitting; one correct and one overfitting example patch in context; temperature 0; CL4SE (context learning for software engineering); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 Max | 80.41 | #201 |
| 2 | GPT-OSS-120B | 76.19 | #330 |
| 3 | Claude 3.5 Haiku (20241022) | 74.26 | #574 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=cl4se-patch-correctness-assessment-2-shot · How It Works · Data refreshed daily, snapshot 2026-10-11.