Knowledge Enhanced Attention for Robust Natural Language Inference
This addresses robustness issues in NLI models for NLP applications, but it is incremental as it builds on existing attention mechanisms and models.
The paper tackles the problem of robustness in natural language inference models by incorporating external knowledge into the attention mechanism, showing that this method significantly improves robustness and achieves human-level performance on the adversarial SNLI dataset when combined with BERT pretraining.
Neural network models have been very successful at achieving high accuracy on natural language inference (NLI) tasks. However, as demonstrated in recent literature, when tested on some simple adversarial examples, most of the models suffer a significant drop in performance. This raises the concern about the robustness of NLI models. In this paper, we propose to make NLI models robust by incorporating external knowledge to the attention mechanism using a simple transformation. We apply the new attention to two popular types of NLI models: one is Transformer encoder, and the other is a decomposable model, and show that our method can significantly improve their robustness. Moreover, when combined with BERT pretraining, our method achieves the human-level performance on the adversarial SNLI data set.