Question Answering on GPQA Diamond (Pass@20)
83.84Pass@20Entropy-Tree
Evaluation Results
| Method | Links | |
|---|---|---|
| Entropy-TreeModel=Qwen2.5-7B-Instruct2026.01 | 83.84 | |
| Entropy-TreeModel=Qwen2.5-32B-Instruct2026.01 | 83.84 | |
| Multi-chainModel=Qwen2.5-32B-Instruct2026.01 | 82.83 | |
| Entropy-TreeModel=Qwen2.5-14B-Instruct2026.01 | 79.8 | |
| Multi-chainModel=Qwen2.5-7B-Instruct2026.01 | 78.79 | |
| Multi-chainModel=Qwen2.5-14B-Instruct2026.01 | 78.28 |