Stacey, J; Minervini, P; Dubossarsky, H; Riedel, S; Rocktäschel, T; (2020) Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training. In: Webber, B and Cohn, T and He, Y and Liu, Y, (eds.) Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). (pp. pp. 8281-8291). Association for Computational Linguistics: Stroudsburg, PA, USA.

Preview

Text 2020.emnlp-main.665.pdf - Published Version Download (390kB) | Preview

Abstract

Natural Language Inference (NLI) datasets contain annotation artefacts resulting in spurious correlations between the natural language utterances and their respective entailment classes. These artefacts are exploited by neural networks even when only considering the hypothesis and ignoring the premise, leading to unwanted biases. Belinkov et al. (2019b) proposed tackling this problem via adversarial training, but this can lead to learned sentence representations that still suffer from the same biases. We show that the bias can be reduced in the sentence representations by using an ensemble of adversaries, encouraging the model to jointly decrease the accuracy of these different adversaries while fitting the data. This approach produces more robust NLI models, outperforming previous de-biasing efforts when generalised to 12 other NLI datasets (Belinkov et al., 2019a; Mahabadi et al., 2020). In addition, we find that the optimal number of adversarial classifiers depends on the dimensionality of the sentence representations, with larger sentence representations being more difficult to de-bias while benefiting from using a greater number of adversaries.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLJSONCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item