HR@JUST Team at SemEval-2020 Task 4: The Impact of RoBERTa Transformer for Evaluation Common Sense Understanding

Heba Al-Jarrah, Rahaf Al-Hamouri, Mohammad AL-Smadi

Abstract
This paper describes the results of our team HR@JUST participation at SemEval-2020 Task 4 - Commonsense Validation and Explanation (ComVE) for POST evaluation period. The provided task consists of three sub-tasks, we participate in task A. We considered a state-of-the-art approach for solving this task by performing RoBERTa model with no Next Sentences Prediction (NSP), dynamic masking, larger training data, and larger batch size. The achieved results show that we got the 11th rank on the final test set leaderboard with an accuracy of 91.3%.
Anthology ID:
2020.semeval-1.64
Volume:
Proceedings of the Fourteenth Workshop on Semantic Evaluation
Month:
December
Year:
2020
Address:
Barcelona (online)
Venue:
SemEval
SIG:
SIGLEX
Publisher:
International Committee for Computational Linguistics
Note:
Pages:
521–526
Language:
URL:
https://aclanthology.org/2020.semeval-1.64
DOI:
10.18653/v1/2020.semeval-1.64
Bibkey:
Cite (ACL):
Heba Al-Jarrah, Rahaf Al-Hamouri, and Mohammad AL-Smadi. 2020. HR@JUST Team at SemEval-2020 Task 4: The Impact of RoBERTa Transformer for Evaluation Common Sense Understanding. In Proceedings of the Fourteenth Workshop on Semantic Evaluation, pages 521–526, Barcelona (online). International Committee for Computational Linguistics.
Cite (Informal):
HR@JUST Team at SemEval-2020 Task 4: The Impact of RoBERTa Transformer for Evaluation Common Sense Understanding (Al-Jarrah et al., SemEval 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/paclic-22-ingestion/2020.semeval-1.64.pdf
PDF Search