ExBERT: An External Knowledge Enhanced BERT for Natural Language Inference

TL;DR

ExBERT enhances BERT with external commonsense knowledge sources to improve natural language inference, achieving state-of-the-art results on SciTail and SNLI benchmarks.

Contribution

The paper introduces ExBERT, a novel model that integrates external knowledge into BERT for improved reasoning in NLI tasks.

Findings

ExBERT achieves 95.9% accuracy on SciTail.

ExBERT achieves 91.5% accuracy on SNLI.

External knowledge integration improves NLI performance.

Abstract

Neural language representation models such as BERT, pre-trained on large-scale unstructured corpora lack explicit grounding to real-world commonsense knowledge and are often unable to remember facts required for reasoning and inference. Natural Language Inference (NLI) is a challenging reasoning task that relies on common human understanding of language and real-world commonsense knowledge. We introduce a new model for NLI called External Knowledge Enhanced BERT (ExBERT), to enrich the contextual representation with real-world commonsense knowledge from external knowledge sources and enhance BERT's language understanding and reasoning capabilities. ExBERT takes full advantage of contextual word representations obtained from BERT and employs them to retrieve relevant external knowledge from knowledge graphs and to encode the retrieved external knowledge. Our model adaptively incorporates the external knowledge context required for reasoning over the inputs. Extensive experiments on the challenging SciTail and SNLI benchmarks demonstrate the effectiveness of ExBERT: in comparison to the previous state-of-the-art, we obtain an accuracy of 95.9% on SciTail and 91.5% on SNLI.