Empowering Language Understanding with Counterfactual Reasoning
This work addresses the limitation of existing language understanding methods that lack counterfactual reasoning, potentially enhancing robustness for NLP applications, though it appears incremental as it builds on existing paradigms with a novel twist.
The paper tackles the problem of language understanding by introducing a Counterfactual Reasoning Model that mimics human counterfactual thinking to improve performance on hard testing samples, achieving validated effectiveness in sentiment analysis and natural language inference experiments.
Present language understanding methods have demonstrated extraordinary ability of recognizing patterns in texts via machine learning. However, existing methods indiscriminately use the recognized patterns in the testing phase that is inherently different from us humans who have counterfactual thinking, e.g., to scrutinize for the hard testing samples. Inspired by this, we propose a Counterfactual Reasoning Model, which mimics the counterfactual thinking by learning from few counterfactual samples. In particular, we devise a generation module to generate representative counterfactual samples for each factual sample, and a retrospective module to retrospect the model prediction by comparing the counterfactual and factual samples. Extensive experiments on sentiment analysis (SA) and natural language inference (NLI) validate the effectiveness of our method.