Logical Reasoning on ReClor v1 (test-e)
85.2AccuracyAPOLLO
Evaluation Results
| Method | Links | |
|---|---|---|
| APOLLOBase Architecture=DeBERTa-v2-xxlarge2022.12 | 85.2 | |
| MERITBase Architecture=DeBERTa-v2-xxlarge2022.12 | 84.6 | |
| DeBERTa-v2-xxlargeBase Architecture=DeBERTa-v2-xxlarge2022.12 | 84 | |
| APOLLOBase Architecture=DeBERTa-v32022.12 | 81.8 | |
| DeBERTa-v3Base Architecture=DeBERTa-v32022.12 | 80.2 |