Logical Reasoning on ReClor v1 (dev)
81.8AccuracyAPOLLO
Evaluation Results
| Method | Links | |
|---|---|---|
| APOLLOBase Architecture=DeBERTa-v2-xxlarge2022.12 | 81.8 | |
| MERITBase Architecture=DeBERTa-v2-xxlarge2022.12 | 80.6 | |
| DeBERTa-v2-xxlargeBase Architecture=DeBERTa-v2-xxlarge2022.12 | 78.3 | |
| APOLLOBase Architecture=DeBERTa-v32022.12 | 76.8 | |
| DeBERTa-v3Base Architecture=DeBERTa-v32022.12 | 75.4 |