Mathematical Reasoning on MATH 500 (Avg@2)
82.1Avg@2TTRL
Evaluation Results
| Method | Links | |
|---|---|---|
| TTRLBase=Base, Verifier=Rule2026.01 | 82.1 | |
| Oat-ZeroBase=Math, Verifier=Rule2026.01 | 80.8 | |
| RLPRBase=Base, Verifier=None2026.01 | 78 | |
| SimpleRL-ZooBase=Math, Verifier=Rule2026.01 | 77.1 | |
| General ReasonerBase=Base, Verifier=Model2026.01 | 77 | |
| DARLBase=Base, Verifier=None2026.01 | 76.6 | |
| RLVRBase=Base, Verifier=Rule2026.01 | 76.5 | |
| PRIMEBase=Math, Verifier=Rule2026.01 | 76.4 | |
| SimpleRL-ZooBase=Base, Verifier=Rule2026.01 | 76.3 | |
| Qwen2.5-7B-InstBase=-, Verifier=None2026.01 | 75.4 | |
| VeriFreeBase=Base, Verifier=None2026.01 | 73.5 | |
| Qwen2.5-7BBase=-, Verifier=None2026.01 | 63 | |
| RLPRBase=Inst, Verifier=None2026.01 | 54.1 | |
| DARLBase=Inst, Verifier=None2026.01 | 52.8 | |
| RLVRBase=Inst, Verifier=Rule2026.01 | 51.9 | |
| Llama3.1-8B-InstBase=-, Verifier=None2026.01 | 50.1 |