Mathematical Reasoning on HLE Math-text (Pass@1)
62.8Pass@1RSE
Evaluation Results
| Method | Links | |
|---|---|---|
| RSEBackbone=Deepseek-V3.2, Iteration=It22026.01 | 62.8 | |
| RSEBackbone=Deepseek-V3.2, Iteration=It12026.01 | 59.2 | |
| MV@128Backbone=Deepseek-V3.2, Iteration=It02026.01 | 58.4 | |
| PaCoReBackbone=Deepseek-V3.2, Iteration=It12026.01 | 58.1 | |
| PaCoReBackbone=Deepseek-V3.2, Iteration=It22026.01 | 57.9 | |
| BaseBackbone=Deepseek-V3.2, Iteration=It02026.01 | 49.3 | |
| PaCoReBackbone=Deepseek-V3.2, Iteration=It02026.01 | 49.3 | |
| RSEBackbone=Deepseek-V3.2, Iteration=It02026.01 | 49.3 | |
| RSEBackbone=Qwen3-30B-A3B-Thinking-2507, Inference Strategy=Recycling Search Experience, Iteration=It32026.01 | 44.8 | |
| BaseBackbone=Qwen3-30B-A3B-Thinking-2507, Inference Strategy=Standard Sampling, Iteration=It02026.01 | 24 | |
| RSEBackbone=Phi-4-Reasoning, Inference Strategy=Recycling Search Experience, Iteration=It32026.01 | 11.4 | |
| BaseBackbone=Phi-4-Reasoning, Inference Strategy=Standard Sampling, Iteration=It02026.01 | 8.5 |