Semi-supervised News Discourse Profiling with Contrastive Learning

TL;DR

This paper introduces a semi-supervised method using contrastive learning for news discourse profiling, effectively addressing data scarcity and improving sentence categorization in news articles.

Contribution

It presents the first semi-supervised approach for news discourse profiling, leveraging intra-document contrastive learning with distillation to utilize structural cues.

Findings

The proposed method outperforms previous supervised approaches.

Semi-supervised learning reduces the need for extensive annotated data.

Evaluation confirms the effectiveness of the contrastive learning framework.

Abstract

News Discourse Profiling seeks to scrutinize the event-related role of each sentence in a news article and has been proven useful across various downstream applications. Specifically, within the context of a given news discourse, each sentence is assigned to a pre-defined category contingent upon its depiction of the news event structure. However, existing approaches suffer from an inadequacy of available human-annotated data, due to the laborious and time-intensive nature of generating discourse-level annotations. In this paper, we present a novel approach, denoted as Intra-document Contrastive Learning with Distillation (ICLD), for addressing the news discourse profiling task, capitalizing on its unique structural characteristics. Notably, we are the first to apply a semi-supervised methodology within this task paradigm, and evaluation demonstrates the effectiveness of the presented approach.