HellaSwag: Can a Machine Really Finish Your Sentence?

GPTKB entity

Statements (30)
Predicate Object
gptkbp:instanceOf gptkb:performance
gptkb:dataset
gptkbp:citation 1000+
gptkbp:contains context sentence
four possible endings
gptkbp:creator gptkb:Yejin_Choi
gptkb:Ali_Farhadi
gptkb:Rowan_Zellers
Ari Holtzman
Yonatan Bisk
gptkbp:difficulty harder than SWAG
gptkbp:focus commonsense story completion
gptkbp:format multiple choice
gptkbp:goal choose the most plausible ending
https://www.w3.org/2000/01/rdf-schema#label HellaSwag: Can a Machine Really Finish Your Sentence?
gptkbp:language English
gptkbp:license gptkb:MIT_License
gptkbp:memiliki_tugas commonsense reasoning
sentence completion
gptkbp:notableFor adversarial filtering
challenging for language models
gptkbp:numberOfRooms 49,000
gptkbp:publicationYear 2019
gptkbp:publishedIn Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019)
gptkbp:relatedTo SWAG dataset
gptkbp:url https://rowanzellers.com/hellaswag/
https://arxiv.org/abs/1905.07830
gptkbp:usedFor evaluating natural language inference models
gptkbp:bfsParent gptkb:HellaSwag
gptkbp:bfsLayer 8