Mathematical Reasoning on MATH (Exact Match)
76.8Exact MatchDarkForest
Evaluation results
| Method | Links | |
|---|---|---|
| DarkForestModels=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 76.8 | |
| Self-ConsistencyModels=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 71.8 | |
| Graph-of-Agent (Mean)Models=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 71.4 | |
| Mixture-of-AgentModels=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 67 | |
| Graph-of-Agent (Max)Models=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 61.4 | |
| DebateModels=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 57.4 | |
| RefineModels=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 51.6 | |
| ReConcileModels=Qwen2.5-7B-Instruct, Qwen2.5-Coder-7B-Instruct, Mathstral-7B-v0.12026.05 | 27.2 |