Mathematical Reasoning on AMC23 (Length and Retention Analysis)
10.9Full LengthMinimal-core extraction
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Minimal-core extractionModel=GPT-52026.05 | 10.9 | 5.2 | 48 | 52 | 69 | 90 | |
| Minimal-core extractionModel=DeepSeek-R1-Distill-Qwen-32B2026.05 | 10.6 | 5.5 | 52 | 48 | 65 | 87 | |
| Minimal-core extractionModel=Qwen3-32B2026.05 | 10 | 5.7 | 57 | 43 | 61 | 85 | |
| Minimal-core extractionModel=DeepSeek-R1-Distill-Qwen-7B2026.05 | 9.8 | 6 | 61 | 39 | 58 | 81 |