Self-Supervised and Controlled Multi-Document Opinion Summarization

TL;DR

This paper introduces a self-supervised, controllable multi-document summarization method that improves summary relevance and coherence for user reviews using an extended Transformer architecture and control codes, outperforming existing models.

Contribution

It presents a novel self-supervised framework with control mechanisms for multi-document summarization, enhancing relevance and reducing hallucinations compared to prior approaches.

Findings

Outperforms graph-based and neural unsupervised models in quality and relevance

Human evaluation confirms improved faithfulness of summaries

Ablation study highlights the effectiveness of control codes in steering summaries

Abstract

We address the problem of unsupervised abstractive summarization of collections of user generated reviews with self-supervision and control. We propose a self-supervised setup that considers an individual document as a target summary for a set of similar documents. This setting makes training simpler than previous approaches by relying only on standard log-likelihood loss. We address the problem of hallucinations through the use of control codes, to steer the generation towards more coherent and relevant summaries.Finally, we extend the Transformer architecture to allow for multiple reviews as input. Our benchmarks on two datasets against graph-based and recent neural abstractive unsupervised models show that our proposed method generates summaries with a superior quality and relevance.This is confirmed in our human evaluation which focuses explicitly on the faithfulness of generated summaries We also provide an ablation study, which shows the importance of the control setup in controlling hallucinations and achieve high sentiment and topic alignment of the summaries with the input reviews.