Improving Zero-Shot Multi-Lingual Entity Linking

TL;DR

This paper introduces a multilingual neural ranker for entity linking that improves zero-shot transfer across languages by learning language-invariant representations, enhancing recall and performance in unseen languages.

Contribution

The paper proposes a neural ranker architecture with adversarial training to improve zero-shot multilingual entity linking performance.

Findings

Adversarial training improves recall in unseen languages.

Zero-shot transfer causes a performance drop that can be mitigated.

Multilingual transformer representations facilitate cross-language linking.

Abstract

Entity linking -- the task of identifying references in free text to relevant knowledge base representations -- often focuses on single languages. We consider multilingual entity linking, where a single model is trained to link references to same-language knowledge bases in several languages. We propose a neural ranker architecture, which leverages multilingual transformer representations of text to be easily applied to a multilingual setting. We then explore how a neural ranker trained in one language (e.g. English) transfers to an unseen language (e.g. Chinese), and find that while there is a consistent but not large drop in performance. How can this drop in performance be alleviated? We explore adding an adversarial objective to force our model to learn language-invariant representations. We find that using this approach improves recall in several datasets, often matching the in-language performance, thus alleviating some of the performance loss occurring from zero-shot transfer.