A Pilot Study of Domain Adaptation Effect for Neural Abstractive Summarization

TL;DR

This study investigates how neural abstractive summarization models transfer knowledge across domains, highlighting the benefits of pre-training and combined data setups for improved summarization quality.

Contribution

It provides initial insights into domain transferability for neural summarization and demonstrates the importance of in-domain data for style adaptation.

Findings

Pre-training on extractive summaries improves neural summarization.

Combining in-domain and out-of-domain data enhances results when in-domain data is limited.

Models can identify salient content across domains but need in-domain data for style capture.

Abstract

We study the problem of domain adaptation for neural abstractive summarization. We make initial efforts in investigating what information can be transferred to a new domain. Experimental results on news stories and opinion articles indicate that neural summarization model benefits from pre-training based on extractive summaries. We also find that the combination of in-domain and out-of-domain setup yields better summaries when in-domain data is insufficient. Further analysis shows that, the model is capable to select salient content even trained on out-of-domain data, but requires in-domain data to capture the style for a target domain.