Logical Reasoning on FOLIO full expert-curated
79.9AccuracyLogicAgent
Evaluation Results
| Method | Links | |
|---|---|---|
| LogicAgentType=AR+SR, Base Model=Qwen2.52025.09 | 79.9 | |
| ToTType=AR, Base Model=Qwen2.52025.09 | 72.06 | |
| Logic-LMType=SR, Base Model=Qwen2.52025.09 | 71.93 | |
| CRType=AR, Base Model=Qwen2.52025.09 | 71.57 | |
| SymbCoTType=SR, Base Model=Qwen2.52025.09 | 70.59 | |
| AristotleType=AR+SR, Base Model=Qwen2.52025.09 | 68.68 | |
| CoTType=LR, Base Model=Qwen2.52025.09 | 68.42 | |
| NaiveType=LR, Base Model=Qwen2.52025.09 | 60.29 |