Error Recognition on Arc
80.4Top-1 AccuracyCORRECT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CORRECTSynthesized by=GPT-4o-mini2025.09 | 80.4 | 88.2 | 91.2 | |
| CORRECTSynthesized by=GPT-5-Nano2025.09 | 80.4 | 88.2 | 91.2 | |
| LLM-as-a-JudgeSynthesized by=GPT-4o-mini2025.09 | 64 | 75 | 78 | |
| BaselineSynthesized by=GPT-5-Nano2025.09 | 62.8 | 69.6 | 71.1 |