Improving Implicit Discourse Relation Recognition with Natural Language Explanations from LLMs

TL;DR

This paper introduces a method that leverages large language models to generate explanations for implicit discourse relations, improving both the accuracy and interpretability of recognition models across multiple NLP tasks.

Contribution

The authors propose a novel classification-generation framework that distills LLM reasoning into lightweight models, enhancing performance and interpretability in implicit discourse relation recognition.

Findings

Significant performance improvement on PDTB dataset

Generated explanations increase model interpretability

Method generalizes to sentiment classification and NLI

Abstract

Implicit Discourse Relation Recognition (IDRR) remains a challenging task due to the requirement for deep semantic understanding in the absence of explicit discourse markers. A further limitation is that existing methods only predict relations without providing any supporting explanations. Recent advances in large language models (LLMs) have shown strong reasoning capabilities in both deep language understanding and natural language explanation generation. In this work, we propose a simple yet effective approach to distill the reasoning capabilities of LLMs into lightweight IDRR models to improve both performance and interpretability. Specifically, we first prompt an LLM to generate explanations for each training instance conditioned on its gold label. Then, we introduce a novel classification-generation framework that jointly performs relation prediction and explanation generation, and train it with the additional supervision of LLM-generated explanations. Our framework is plug-and-play, enabling easy integration with most existing IDRR models. Experimental results on PDTB demonstrate that our approach significantly improves IDRR performance, while human evaluation further confirms that the generated explanations enhance model interpretability. Furthermore, we validate the generality of our approach on sentiment classification and natural language inference