Unsupervised Extractive Summarization using Pointwise Mutual Information

TL;DR

This paper introduces an unsupervised extractive summarization method that uses pointwise mutual information (PMI) to measure sentence relevance and redundancy, leading to improved summaries across diverse domains.

Contribution

It proposes novel PMI-based metrics for relevance and redundancy, and a greedy algorithm for sentence selection, outperforming traditional similarity-based methods.

Findings

Outperforms similarity-based methods on multiple datasets

Effective across news, medical, and personal texts

Uses pre-trained language models for PMI computation

Abstract

Unsupervised approaches to extractive summarization usually rely on a notion of sentence importance defined by the semantic similarity between a sentence and the document. We propose new metrics of relevance and redundancy using pointwise mutual information (PMI) between sentences, which can be easily computed by a pre-trained language model. Intuitively, a relevant sentence allows readers to infer the document content (high PMI with the document), and a redundant sentence can be inferred from the summary (high PMI with the summary). We then develop a greedy sentence selection algorithm to maximize relevance and minimize redundancy of extracted sentences. We show that our method outperforms similarity-based methods on datasets in a range of domains including news, medical journal articles, and personal anecdotes.