OpineSum: Entailment-based self-training for abstractive opinion summarization

TL;DR

OpineSum introduces a novel entailment-based self-training method for abstractive opinion summarization, effectively capturing consensus across reviews and achieving state-of-the-art results in unsupervised and few-shot settings.

Contribution

The paper presents a new self-training approach using textual entailment for opinion summarization, enabling large-scale silver-standard summary generation and improved performance.

Findings

Achieves state-of-the-art results in opinion summarization

Effectively captures consensus across multiple reviews

Enables large-scale unsupervised and few-shot training

Abstract

A typical product or place often has hundreds of reviews, and summarization of these texts is an important and challenging problem. Recent progress on abstractive summarization in domains such as news has been driven by supervised systems trained on hundreds of thousands of news articles paired with human-written summaries. However for opinion texts, such large scale datasets are rarely available. Unsupervised methods, self-training, and few-shot learning approaches bridge that gap. In this work, we present a novel self-training approach, OpineSum, for abstractive opinion summarization. The summaries in this approach are built using a novel application of textual entailment and capture the consensus of opinions across the various reviews for an item. This method can be used to obtain silver-standard summaries on a large scale and train both unsupervised and few-shot abstractive summarization systems. OpineSum achieves state-of-the-art performance in both settings.