Mathematical Reasoning on HMMT25 (Pass@1)
20Pass@1Qwen3-4B +CAST
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-4B +CASTModel Scale=4B, Method=+CAST2026.05 | 20 | |
| Qwen3-8B +CASTModel Scale=8B, Method=+CAST2026.05 | 20 | |
| Qwen3-1.7B +CASTModel Scale=1.7B, Method=+CAST2026.05 | 13.3 | |
| Qwen3-4B +GRPOModel Scale=4B, Method=+GRPO2026.05 | 13.3 | |
| Qwen3-4B +GRPO+OPSDModel Scale=4B, Method=+GRPO+OPSD2026.05 | 13.3 | |
| Qwen3-4B +RLSDModel Scale=4B, Method=+RLSD2026.05 | 13.3 | |
| Qwen3-4B +RLRTModel Scale=4B, Method=+RLRT2026.05 | 13.3 | |
| Qwen3-1.7B +RLSDModel Scale=1.7B, Method=+RLSD2026.05 | 10 | |
| Qwen3-4B BaseModel Scale=4B, Method=Base2026.05 | 10 | |
| Qwen3-8B BaseModel Scale=8B, Method=Base2026.05 | 10 | |
| Qwen3-8B +GRPO+OPSDModel Scale=8B, Method=+GRPO+OPSD2026.05 | 10 | |
| Qwen3-1.7B BaseModel Scale=1.7B, Method=Base2026.05 | 6.7 | |
| Qwen3-1.7B +GRPO+OPSDModel Scale=1.7B, Method=+GRPO+OPSD2026.05 | 6.7 | |
| Qwen3-1.7B +RLRTModel Scale=1.7B, Method=+RLRT2026.05 | 6.7 | |
| Qwen3-8B +GRPOModel Scale=8B, Method=+GRPO2026.05 | 6.7 | |
| Qwen3-8B +OPSDModel Scale=8B, Method=+OPSD2026.05 | 6.7 | |
| Qwen3-8B +RLSDModel Scale=8B, Method=+RLSD2026.05 | 6.7 | |
| Qwen3-8B +RLRTModel Scale=8B, Method=+RLRT2026.05 | 6.7 | |
| Qwen3-1.7B +GRPOModel Scale=1.7B, Method=+GRPO2026.05 | 3.3 | |
| Qwen3-1.7B +OPSDModel Scale=1.7B, Method=+OPSD2026.05 | 0 | |
| Qwen3-4B +OPSDModel Scale=4B, Method=+OPSD2026.05 | 0 |