Advanced Mathematical Reasoning on Math500 (512 tokens)
46.3Pass@1 Accuracyd-TreeRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| d-TreeRPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 46.3 | |
| TraceRLBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 44.1 | |
| Diffu-GRPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 43.4 | |
| wd1Base Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 43.4 | |
| LLaDA-MoE-7BA1B-InstructBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 42.2 | |
| GDPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 41.2 | |
| SAPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 40.4 | |
| TraceRLBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 39.1 | |
| Diffu-GRPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 39 | |
| wd1Base Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 39 | |
| d-TreeRPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 38.9 | |
| GDPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 38.5 | |
| SAPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 38.4 | |
| LLaDA-8B-InstructBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 36.2 | |
| VRPO (LLaDA-1.5)Base Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 34.8 |