Models and Datasets for Cross-Lingual Summarisation

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 9408–9423 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 9408 Models and Datasets for Cross-Lingual Summarisation Laura Perez-Beltrachini and Mirella Lapata Institute for Language, Cognition and Computation School of Informatics, University of Edinburgh 10 Crichton Street, Edinburgh EH8 9AB {lperez,mlap}@inf.ed.ac.uk Abstract We present a cross-lingual summarisation cor- pus with long documents in a source language associated with multi-sentence summaries in a target language. The corpus covers twelve lan- guage pairs and directions for four European languages, namely Czech, English, French and German, and the methodology for its cre- ation can be applied to several other languages. We derive cross-lingual document-summary instances from Wikipedia by combining lead paragraphs and articles’ bodies from language aligned Wikipedia titles. We analyse the pro- posed cross-lingual summarisation task with automatic metrics and validate it with a human study. To illustrate the utility of our dataset we report experiments with multi-lingual pre- trained models in supervised, zero- and few- shot, and out-of-domain scenarios. 1 Introduction Given a document in a source language (e.g., French), cross-lingual summarisation aims to pro- duce a summary in a different target language (e.g., English). The practical benefits of this task are twofold: it not only provides rapid access to salient content, but also enables the dissemination of rel- evant content across speakers of other languages. For instance, providing summaries of articles from French or German newspapers to non-French or non-German speakers; or enabling access to sum- mary descriptions of goods, services, or knowledge available online in foreign languages. Figure 1 shows an example of an input document in French (left) and its summary in English and other lan- guages (right). Recent years have witnessed increased interest in abstractive summarisation (Rush et al., 2015; Zhang et al., 2020) thanks to the popularity of neu- ral network models and the availability of datasets (Sandhaus, 2008; Hermann et al., 2015; Grusky et al., 2018) containing hundreds of thousands of document-summary pairs. Although initial efforts have overwhelmingly focused on English, more recently, with the advent of cross-lingual represen- tations (Ruder et al., 2019) and large pre-trained models (Devlin et al., 2019; Liu et al., 2020), re- search on multi-lingual summarisation (i.e., build- ing monolingual summarisation systems for differ- ent languages) has been gaining momentum (Chi et al., 2020b; Scialom et al., 2020). While creating large-scale multi-lingual sum- marisation datasets has proven feasible (Straka et al., 2018; Scialom et al., 2020), at least for the news domain, cross-lingual datasets are more diffi- cult to obtain. In contrast to monolingual sum- marisation, naturally occurring documents in a source language paired with summaries in different target languages are rare. For this reason, exist- ing approaches either create large-scale synthetic data using back-translation (Zhu et al., 2019; Cao et al., 2020), translate the input documents (Ouyang et al., 2019), or build document-summary pairs from social media annotations and crowd-sourcing (Nguyen and Daumé III, 2019). Recent efforts (Ladhak et al., 2020) have been directed at the cre- …

Content truncated to 500 words