A Multilingual Study of Compressive Cross-Language Text Summarization

TL;DR

This paper introduces a compressive framework for cross-language text summarization, demonstrating improved and more stable ROUGE scores across multiple languages compared to existing extractive methods.

Contribution

It proposes a novel compressive approach for CLTS and provides comprehensive multilingual evaluation showing enhanced stability and performance.

Findings

Outperforms state-of-the-art extractive CLTS methods

Achieves higher ROUGE scores across four languages

Demonstrates improved stability in summary quality

Abstract

Cross-Language Text Summarization (CLTS) generates summaries in a language different from the language of the source documents. Recent methods use information from both languages to generate summaries with the most informative sentences. However, these methods have performance that can vary according to languages, which can reduce the quality of summaries. In this paper, we propose a compressive framework to generate cross-language summaries. In order to analyze performance and especially stability, we tested our system and extractive baselines on a dataset available in four languages (English, French, Portuguese, and Spanish) to generate English and French summaries. An automatic evaluation showed that our method outperformed extractive state-of-art CLTS methods with better and more stable ROUGE scores for all languages.