Adversarially Regularising Neural NLI Models to Integrate Logical Background Knowledge

Abstract

Adversarial examples are inputs to machine learning models designed to cause the model to make a mistake. They are useful for understanding the shortcomings of machine learning models, interpreting their results, and for regularisation. In NLP, however, most example generation strategies produce input text by using known, pre-specified semantic transformations, requiring significant manual effort and in-depth understanding of the problem and domain. In this paper, we investigate the problem of automatically generating adversarial examples that violate a set of given First-Order Logic constraints in Natural Language Inference (NLI). We reduce the problem of identifying such adversarial examples to a combinatorial optimisation problem, by maximising a quantity measuring the degree of violation of such constraints and by using a language model for generating linguistically-plausible examples. Furthermore, we propose a method for adversarially regularising neural NLI models for incorporating background knowledge. Our results show that, while the proposed method does not always improve results on the SNLI and MultiNLI datasets, it significantly and consistently increases the predictive accuracy on adversarially-crafted datasets -- up to a 79.6% relative improvement -- while drastically reducing the number of background knowledge violations. Furthermore, we show that adversarial examples transfer among model architectures, and that the proposed adversarial training procedure improves the robustness of NLI models to adversarial examples.

Accepted at the SIGNLL Conference on Computational Natural Language Learning (CoNLL 2018)

References in corpus (5)

• Explaining and Harnessing Adversarial Examples
• Synthetic and Natural Noise Both Break Neural Machine Translation
• Adversarial Examples for Evaluating Reading Comprehension Systems
• Adversarial Evaluation of Dialogue Models
• Generating Sentences by Editing Prototypes

Cited by in corpus (7)

• Deep Learning with Logical Constraints
• Beyond Leaderboards: A survey of methods for revealing weaknesses in Natural Language Inference data and models
• Integrating Deep Learning with Logic Fusion for Information Extraction
• Towards Effective Paraphrasing for Information Disguise
• Augmenting Neural Networks with First-order Logic
• What If We Simply Swap the Two Text Fragments? A Straightforward yet Effective Way to Test the Robustness of Methods to Confounding Signals in Nature Language Inference Tasks
• Predicting Clinical Trial Results by Implicit Evidence Integration