Popular Natural Language Inference (NLI) datasets have been shown to be tainted by hypothesis-only biases. Adversarial learning may help models ignore sensitive biases and spurious correlations in data. We evaluate whether adversarial learning can be used in NLI to encourage models to learn representations free of hypothesis-only biases. Our analyses indicate that the representations learned via adversarial learning may be less biased, with only small drops in NLI accuracy.
@article{arxiv.1907.04389,
title = {On Adversarial Removal of Hypothesis-only Bias in Natural Language Inference},
author = {Yonatan Belinkov and Adam Poliak and Stuart M. Shieber and Benjamin Van Durme and Alexander M. Rush},
journal= {arXiv preprint arXiv:1907.04389},
year = {2019}
}
Comments
StarSem 2019 - The Eighth Joint Conference on Lexical and Computational Semantics