Descartes: Generating Short Descriptions of Wikipedia Articles

TL;DR

This paper introduces Descartes, a multilingual model that automatically generates short descriptions for Wikipedia articles by integrating article text, existing descriptions in other languages, and semantic type info, significantly improving coverage and quality.

Contribution

The paper presents a novel multilingual model for automatic Wikipedia short description generation, outperforming baselines and matching monolingual models in quality, with potential for real-world application.

Findings

Descartes outperforms baseline models including translation-based methods.

The model achieves human-level quality in description generation.

91.3% of Descartes's English descriptions pass Wikipedia inclusion standards.

Abstract

Wikipedia is one of the richest knowledge sources on the Web today. In order to facilitate navigating, searching, and maintaining its content, Wikipedia's guidelines state that all articles should be annotated with a so-called short description indicating the article's topic (e.g., the short description of beer is "Alcoholic drink made from fermented cereal grains"). Nonetheless, a large fraction of articles (ranging from 10.2% in Dutch to 99.7% in Kazakh) have no short description yet, with detrimental effects for millions of Wikipedia users. Motivated by this problem, we introduce the novel task of automatically generating short descriptions for Wikipedia articles and propose Descartes, a multilingual model for tackling it. Descartes integrates three sources of information to generate an article description in a target language: the text of the article in all its language versions, the already-existing descriptions (if any) of the article in other languages, and semantic type information obtained from a knowledge graph. We evaluate a Descartes model trained for handling 25 languages simultaneously, showing that it beats baselines (including a strong translation-based baseline) and performs on par with monolingual models tailored for specific languages. A human evaluation on three languages further shows that the quality of Descartes's descriptions is largely indistinguishable from that of human-written descriptions; e.g., 91.3% of our English descriptions (vs. 92.1% of human-written descriptions) pass the bar for inclusion in Wikipedia, suggesting that Descartes is ready for production, with the potential to support human editors in filling a major gap in today's Wikipedia across languages.