Grade School Math Word Problems on GSM8k (512 tokens)
82.6Pass@1d-TreeRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| d-TreeRPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 82.6 | |
| TraceRLBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 82.4 | |
| wd1Base Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 82.3 | |
| SAPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 82.1 | |
| GDPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 82 | |
| Diffu-GRPOBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 81.9 | |
| VRPO (LLaDA-1.5)Base Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 81.5 | |
| LLaDA-8B-InstructBase Model=LLaDA-8B-Instruct, Generation Length=5122025.12 | 78.2 | |
| d-TreeRPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 77.4 | |
| SAPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 76.8 | |
| wd1Base Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 76.4 | |
| TraceRLBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 76.1 | |
| GDPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 75.2 | |
| Diffu-GRPOBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 73.7 | |
| LLaDA-MoE-7BA1B-InstructBase Model=LLaDA-MoE-7BA1B-Instruct, Generation Length=5122025.12 | 71 |