Scientific Problem Solving on SciBench
77.46Pass@20Entropy-Tree
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Entropy-TreeModel=Qwen2.5-32B-Instruct2026.01 | 77.46 | — | |
| Multi-chainModel=Qwen2.5-14B-Instruct2026.01 | 76.88 | — | |
| Entropy-TreeModel=Qwen2.5-14B-Instruct2026.01 | 76.88 | — | |
| Multi-chainModel=Qwen2.5-32B-Instruct2026.01 | 75.14 | — | |
| Entropy-TreeModel=Qwen2.5-7B-Instruct2026.01 | 66.62 | — | |
| Multi-chainModel=Qwen2.5-7B-Instruct2026.01 | 65.75 | — | |
| CapFlowType=Learning, Setting=Multi-domain, Passes=12026.02 | 41.6 | — | |
| CapFlowType=Learning, Setting=Cross-domain, Passes=1, Cross-domain Domain Mapping=All->Science2026.02 | 41.6 | — | |
| AFlowType=Refinement, Setting=Single-domain, Iterations=202026.02 | 41.4 | — | |
| ScoreFlowType=Learning, Setting=Single-domain, Iterations=52026.02 | 34.2 | — | |
| ScoreFlowType=Learning, Setting=Multi-domain, Iterations=52026.02 | 34.2 | — | |
| ADASType=Refinement, Setting=Single-domain, Iterations=202026.02 | 30.36 | — | |
| GPT-4o-miniType=Manual, Setting=Manual2026.02 | 28.4 | — | |
| SPPType=Manual, Setting=Manual2026.02 | 28.22 | — | |
| CoT-SCType=Manual, Setting=Manual2026.02 | 27.27 | — | |
| Self-RefineType=Manual, Setting=Manual2026.02 | 27.03 | — | |
| CoTType=Manual, Setting=Manual2026.02 | 24.83 | — |