Summarization with Precise Length Control

TL;DR

This paper introduces a new framework for length-controlled summarization that generates summaries with precise length specifications and improves overall quality by jointly predicting lengths, evaluated on the CNNDM dataset.

Contribution

The paper proposes a novel framework that achieves exact length control in summarization and jointly trains models to predict summary lengths, enhancing performance.

Findings

Improved length accuracy in summaries.

Enhanced summarization quality compared to existing methods.

Effective length prediction integrated into the generation process.

Abstract

Many applications of text generation such as summarization benefit from accurately controlling the text length. Existing approaches on length-controlled summarization either result in degraded performance or can only control the length approximately. In this work, we present a framework to generate summaries with precisely the specified number of tokens or sentences, while maintaining or even improving the text quality. In addition, we jointly train the models to predict the lengths, so our model can generate summaries with optimal length. We evaluate the proposed framework on the CNNDM dataset and show improved performance compared to existing methods.