Robustness to Modification with Shared Words in Paraphrase Identification

Abstract

Revealing the robustness issues of natural language processing models and improving their robustness is important to their performance under difficult situations. In this paper, we study the robustness of paraphrase identification models from a new perspective -- via modification with shared words, and we show that the models have significant robustness issues when facing such modifications. To modify an example consisting of a sentence pair, we either replace some words shared by both sentences or introduce new shared words. We aim to construct a valid new example such that a target model makes a wrong prediction. To find a modification solution, we use beam search constrained by heuristic rules, and we leverage a BERT masked language model for generating substitution words compatible with the context. Experiments show that the performance of the target models has a dramatic drop on the modified examples, thereby revealing the robustness issue. We also show that adversarial training can mitigate this issue.

Findings of EMNLP 2020

References in corpus (8)

• Convolutional Neural Network Architectures for Matching Natural Language Sentences
• Natural Language Inference over Interaction Space
• Towards Crafting Text Adversarial Samples
• Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment
• Adversarial Texts with Gradient Methods
• Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information
• Greedy Attack and Gumbel Attack: Generating Adversarial Examples for Discrete Data
• Robust Machine Comprehension Models via Adversarial Training