On the Trade-off between Redundancy and Local Coherence in Summarization

TL;DR

This paper explores how controlling for redundancy and cohesion affects the quality of extractive summaries, especially for long, redundant documents, using both supervised and unsupervised optimization methods inspired by cognitive theories.

Contribution

It introduces two unsupervised systems that incorporate psycholinguistic principles to balance informativeness, redundancy, and cohesion in summaries, demonstrating their effectiveness through evaluations.

Findings

Systems optimizing for cohesion produce more organized summaries.

Unsupervised systems achieve high cohesion but with reduced informativeness.

Cognitive-inspired models influence the trade-off between summary properties.

Abstract

Extractive summaries are usually presented as lists of sentences with no expected cohesion between them and with plenty of redundant information if not accounted for. In this paper, we investigate the trade-offs incurred when aiming to control for inter-sentential cohesion and redundancy in extracted summaries, and their impact on their informativeness. As case study, we focus on the summarization of long, highly redundant documents and consider two optimization scenarios, reward-guided and with no supervision. In the reward-guided scenario, we compare systems that control for redundancy and cohesion during sentence scoring. In the unsupervised scenario, we introduce two systems that aim to control all three properties -- informativeness, redundancy, and cohesion -- in a principled way. Both systems implement a psycholinguistic theory that simulates how humans keep track of relevant content units and how cohesion and non-redundancy constraints are applied in short-term memory during reading. Extensive automatic and human evaluations reveal that systems optimizing for -- among other properties -- cohesion are capable of better organizing content in summaries compared to systems that optimize only for redundancy, while maintaining comparable informativeness. We find that the proposed unsupervised systems manage to extract highly cohesive summaries across varying levels of document redundancy, although sacrificing informativeness in the process. Finally, we lay evidence as to how simulated cognitive processes impact the trade-off between the analyzed summary properties.