HellaSwag

Mosaic • 2019
HellaSWAG is a dataset for studying grounded commonsense inference. It consists of 70k multiple choice questions about grounded situations: each question comes from one of two domains -- activitynet or wikihow -- with four answer choices about what might happen next in the scene. The correct answer is the (real) sentence for the next event; the three incorrect answers are adversarially generated and human verified, so as to fool machines but not humans.
License: MIT

Leaderboard

Top Public Submissions
DetailsCreatedAccuracy
1
CompassMTL
Anonymous
5/11/202296%
2
DeBERTa Large
EMNLP Paper 3842 Authors
5/20/202296%
3
CreAT
Anonymous
5/3/202295%
4
DeBERTa MCQ
EMNLP Paper 3842 Authors
6/3/202295%
5
DeBERTa Large
Anonymous
4/14/202294%