Prompt-based Connective Prediction Method for Fine-grained Implicit Discourse Relation Recognition

TL;DR

This paper introduces a prompt-based connective prediction method that leverages large-scale pre-trained models to improve fine-grained implicit discourse relation recognition, significantly outperforming existing models and transferring effectively to explicit discourse relation recognition.

Contribution

The paper proposes a novel prompt-based approach that utilizes discourse knowledge and connectives to enhance implicit discourse relation recognition, especially in few-shot scenarios.

Findings

Surpasses state-of-the-art in fine-grained IDRR

Achieves significant improvements on few-shot discourse relations

Effectively transfers to explicit discourse relation recognition

Abstract

Due to the absence of connectives, implicit discourse relation recognition (IDRR) is still a challenging and crucial task in discourse analysis. Most of the current work adopted multi-task learning to aid IDRR through explicit discourse relation recognition (EDRR) or utilized dependencies between discourse relation labels to constrain model predictions. But these methods still performed poorly on fine-grained IDRR and even utterly misidentified on most of the few-shot discourse relation classes. To address these problems, we propose a novel Prompt-based Connective Prediction (PCP) method for IDRR. Our method instructs large-scale pre-trained models to use knowledge relevant to discourse relation and utilizes the strong correlation between connectives and discourse relation to help the model recognize implicit discourse relations. Experimental results show that our method surpasses the current state-of-the-art model and achieves significant improvements on those fine-grained few-shot discourse relation. Moreover, our approach is able to be transferred to EDRR and obtain acceptable results. Our code is released in https://github.com/zh-i9/PCP-for-IDRR.