HellaSwag is a challenge dataset for evaluating commonsense NLI that is specially hard for state-of-the-art models, though its questions are trivial for humans (>95% accuracy).
Benchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
HellaSwag
|
CompassMTL 567M with Tailor
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
