Original paper
A Surprisingly Robust Trick for the Winograd Schema Challenge
Pages: 4837 - 4842
Published: Jan 1, 2019
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...
Paper Details
Title
A Surprisingly Robust Trick for the Winograd Schema Challenge
Published Date
Jan 1, 2019
Pages
4837 - 4842