Structured Reasoning on HARDMath
67.3Score (%)Prompt
Evaluation results
| Method | Links | |
|---|---|---|
| PromptModel=Qwen3-30B-A3B, Setting=CoT overall2026.05 | 67.3 | |
| Memory InceptionModel=Qwen3-30B-A3B, Setting=CoT overall2026.05 | 63.8 | |
| PlainModel=Qwen3-30B-A3B, Setting=CoT overall2026.05 | 60.8 |