A Surprisingly Robust Trick for the Winograd Schema Challenge

Vid Kocijan, Ana-Maria Cretu, Oana-Maria Camburu, Yordan Yordanov, Thomas Lukasiewicz


Abstract
The Winograd Schema Challenge (WSC) dataset WSC273 and its inference counterpart WNLI are popular benchmarks for natural language understanding and commonsense reasoning. In this paper, we show that the performance of three language models on WSC273 consistently and robustly improves when fine-tuned on a similar pronoun disambiguation problem dataset (denoted WSCR). We additionally generate a large unsupervised WSC-like dataset. By fine-tuning the BERT language model both on the introduced and on the WSCR dataset, we achieve overall accuracies of 72.5% and 74.7% on WSC273 and WNLI, improving the previous state-of-the-art solutions by 8.8% and 9.6%, respectively. Furthermore, our fine-tuned models are also consistently more accurate on the “complex” subsets of WSC273, introduced by Trichelair et al. (2018).
Anthology ID:
P19-1478
Volume:
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2019
Address:
Florence, Italy
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4837–4842
Language:
URL:
https://aclanthology.org/P19-1478
DOI:
10.18653/v1/P19-1478
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/P19-1478.pdf