Mathematical Reasoning on AMC23 (P@1, ASRt, RAS)
62.5P@1BaseRL
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| BaseRLBackbone=Qwen2.5-14B-Instruct2025.09 | 62.5 | — | — | |
| DecepChainBackbone=Qwen2.5-14B-Instruct2025.09 | 58.69 | 100 | 100 |
| Method | Links | |||
|---|---|---|---|---|
| BaseRLBackbone=Qwen2.5-14B-Instruct2025.09 | 62.5 | — | — | |
| DecepChainBackbone=Qwen2.5-14B-Instruct2025.09 | 58.69 | 100 | 100 |