General Reasoning on GPQA Diamond (Score)
85.4General Reasoning ScoreConfident Decoding
Evaluation Results
| Method | Links | |
|---|---|---|
| Confident DecodingModel=Qwen3.5-122B-A10B, Decoding Strategy=Confident (Ours)2026.06 | 85.4 | |
| Last LayerModel=Qwen3.5-122B-A10B, Decoding Strategy=Last Layer2026.06 | 83.3 | |
| Confident DecodingModel=Qwen3.5-35B-A3B, Decoding Strategy=Confident (Ours)2026.06 | 82.8 | |
| Confident DecodingModel=Gemma-4-31B, Decoding Strategy=Confident (Ours)2026.06 | 80.8 | |
| Confident DecodingModel=gpt-oss-120B, Decoding Strategy=Confident (Ours)2026.06 | 80.8 | |
| Confident DecodingModel=Qwen3.5-27B, Decoding Strategy=Confident (Ours)2026.06 | 79.4 | |
| Last LayerModel=Qwen3.5-27B, Decoding Strategy=Last Layer2026.06 | 78.2 | |
| Last LayerModel=Gemma-4-31B, Decoding Strategy=Last Layer2026.06 | 76.8 | |
| Last LayerModel=Qwen3.5-35B-A3B, Decoding Strategy=Last Layer2026.06 | 76.3 | |
| Last LayerModel=gpt-oss-120B, Decoding Strategy=Last Layer2026.06 | 76.3 | |
| Confident DecodingModel=gpt-oss-20b, Decoding Strategy=Confident (Ours)2026.06 | 60.8 | |
| Last LayerModel=gpt-oss-20b, Decoding Strategy=Last Layer2026.06 | 58.1 |