DYLE: Dynamic Latent Extraction for Abstractive Long-Input Summarization

TL;DR

DYLE introduces a dynamic latent extraction method for abstractive long-input summarization, jointly training an extractor and generator to improve performance on long documents and dialogues with interpretable weights.

Contribution

The paper proposes a novel dynamic latent extraction approach that enhances long-input summarization by jointly training extractor and generator with dynamic attention and supervision heuristics.

Findings

Outperforms existing methods on GovReport and QMSum with up to 6.1 ROUGE improvements.

Effective on long-document and long-dialogue summarization tasks.

Provides interpretable dynamic weights during generation.

Abstract

Transformer-based models have achieved state-of-the-art performance on short-input summarization. However, they still struggle with summarizing longer text. In this paper, we present DYLE, a novel dynamic latent extraction approach for abstractive long-input summarization. DYLE jointly trains an extractor and a generator and treats the extracted text snippets as the latent variable, allowing dynamic snippet-level attention weights during decoding. To provide adequate supervision, we propose simple yet effective heuristics for oracle extraction as well as a consistency loss term, which encourages the extractor to approximate the averaged dynamic weights predicted by the generator. We evaluate our method on different long-document and long-dialogue summarization tasks: GovReport, QMSum, and arXiv. Experiment results show that DYLE outperforms all existing methods on GovReport and QMSum, with gains up to 6.1 ROUGE, while yielding strong results on arXiv. Further analysis shows that the proposed dynamic weights provide interpretability of our generation process.