Informal Logic Reasoning on FOLIO and LogicNLI
64FOLIO ScoreQwen 2.5 7B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Qwen 2.5 7BBackbone=Qwen 2.5 7B, Evaluation Protocol=Best-of-K (K=7), Training=None (Baseline)2025.05 | 64 | 44.4 | 54.2 | |
| FOVER-Qwen2.5-7BBackbone=Qwen 2.5 7B, Evaluation Protocol=Best-of-K (K=7), Training=Fine-tuned on FOVER-40K2025.05 | 63.5 | 49.6 | 56.6 | |
| FOVER-Llama3.1-8BBackbone=Llama 3.1 8B, Evaluation Protocol=Best-of-K (K=7), Training=Fine-tuned on FOVER-40K2025.05 | 60.1 | 41.2 | 50.6 | |
| Llama 3.1 8BBackbone=Llama 3.1 8B, Evaluation Protocol=Best-of-K (K=7), Training=None (Baseline)2025.05 | 58.6 | 39.2 | 48.9 |