Let's be explicit about that: Distant supervision for implicit discourse relation classification via connective prediction

TL;DR

This paper introduces a novel approach to implicit discourse relation classification by leveraging explicitation techniques, transforming the problem into language modeling and explicit relation classification, achieving competitive results without relying on labeled implicit data.

Contribution

The study proposes a new method that sidesteps the need for annotated implicit relations by explicitly modeling and predicting discourse connectives, outperforming or matching state-of-the-art methods.

Findings

Method marginally outperforms state-of-the-art models.

Performance remains robust across different domains.

Language models effectively capture inter-sentence relations without explicit markers.

Abstract

In implicit discourse relation classification, we want to predict the relation between adjacent sentences in the absence of any overt discourse connectives. This is challenging even for humans, leading to shortage of annotated data, a fact that makes the task even more difficult for supervised machine learning approaches. In the current study, we perform implicit discourse relation classification without relying on any labeled implicit relation. We sidestep the lack of data through explicitation of implicit relations to reduce the task to two sub-problems: language modeling and explicit discourse relation classification, a much easier problem. Our experimental results show that this method can even marginally outperform the state-of-the-art, in spite of being much simpler than alternative models of comparable performance. Moreover, we show that the achieved performance is robust across domains as suggested by the zero-shot experiments on a completely different domain. This indicates that recent advances in language modeling have made language models sufficiently good at capturing inter-sentence relations without the help of explicit discourse markers.