Cross-Lingual Transfer in Zero-Shot Cross-Language Entity Linking

TL;DR

This paper introduces a neural ranking model using multilingual BERT for cross-lingual entity linking, demonstrating strong zero-shot transfer capabilities and analyzing factors affecting performance.

Contribution

It presents a novel neural ranking architecture leveraging multilingual BERT and investigates zero-shot transfer, domain shift, and mitigation strategies.

Findings

Multilingual BERT enables robust cross-lingual entity linking.

Zero-shot transfer performance is surprisingly strong.

Auxiliary training reduces zero-shot degradation, but domain shift remains a challenge.

Abstract

Cross-language entity linking grounds mentions in multiple languages to a single-language knowledge base. We propose a neural ranking architecture for this task that uses multilingual BERT representations of the mention and the context in a neural network. We find that the multilingual ability of BERT leads to robust performance in monolingual and multilingual settings. Furthermore, we explore zero-shot language transfer and find surprisingly robust performance. We investigate the zero-shot degradation and find that it can be partially mitigated by a proposed auxiliary training objective, but that the remaining error can best be attributed to domain shift rather than language transfer.