Measure and Improve Robustness in NLP Models: A Survey

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4569 - 4586 July 10-15, 2022 ©2022 Association for Computational Linguistics Measure and Improve Robustness in NLP Models: A Survey Xuezhi Wang Google Research xuezhiw@google.com Haohan Wang Carnegie Mellon University haohanw@cs.cmu.edu Diyi Yang Georgia Institute of Technology dyang888@gatech.edu Abstract As NLP models achieved state-of-the-art per- formances over benchmarks and gained wide applications, it has been increasingly impor- tant to ensure the safe deployment of these models in the real world, e.g., making sure the models are robust against unseen or chal- lenging scenarios. Despite robustness being an increasingly studied topic, it has been sepa- rately explored in applications like vision and NLP, with various definitions, evaluation and mitigation strategies in multiple lines of re- search. In this paper, we aim to provide a uni- fying survey of how to define, measure and improve robustness in NLP. We first connect multiple definitions of robustness, then unify various lines of work on identifying robust- ness failures and evaluating models’ robust- ness. Correspondingly, we present mitigation strategies that are data-driven, model-driven, and inductive-prior-based, with a more system- atic view of how to effectively improve robust- ness in NLP models. Finally, we conclude by outlining open challenges and future direc- tions to motivate further research in this area. 1 Introduction NLP models, especially with the recent advances of large pre-trained language models have achieved great progress and gained wide applications in the real world. Despite the performance gains, NLP models are still fragile and brittle to out-of-domain data (Hendrycks et al., 2020a; Wang et al., 2019d), adversarial attacks (McCoy et al., 2019; Jia and Liang, 2017; Jin et al., 2020), or small perturba- tion to the input (Ebrahimi et al., 2018; Belinkov and Bisk, 2018). Those failures could hinder the safe deployment of these models in the real world, and impact NLP models’ trustworthiness to users. As a result, an increasing line of work has been conducted to understand robustness issues in the language technologies communities. Still, diverse sets of research across multiple dimensions and numerous levels of depth exist and are scattered across various communities; for instance, using a variety of definitions on a wide range of very different NLP tasks. In this work, we provide a uni- fying overview of what is robustness in NLP, how to identify robustness failures and evaluate model’s robustness, and systematic ways to improve robust- ness, as well as a conceptual schema categorizing ongoing research directions. We identify gaps be- tween the to-date robustness work, the technical opportunities, and discuss possible paths forward. 2 Definitions of Robustness in NLP Robustness, despite its specific definitions in var- ious lines of research, can typically be unified as follows: denote the input as x, and its associated gold label for the main task as y, assume a model f is trained on (x, y) ∼ D and its prediction over x as f(x); now given test data (x′, y′) ∼ …

Content truncated to 500 words