Multilingual End to End Entity Linking
This addresses the problem of complex model stacks in multilingual entity linking for NLP applications, though it appears incremental as it builds on existing entity linking tasks.
The paper tackles the lack of efficient end-to-end multilingual entity linking solutions by introducing BELA, the first fully end-to-end model that detects and links entities in 97 languages, achieving performance on four datasets covering high- and low-resource languages.
Entity Linking is one of the most common Natural Language Processing tasks in practical applications, but so far efficient end-to-end solutions with multilingual coverage have been lacking, leading to complex model stacks. To fill this gap, we release and open source BELA, the first fully end-to-end multilingual entity linking model that efficiently detects and links entities in texts in any of 97 languages. We provide here a detailed description of the model and report BELA's performance on four entity linking datasets covering high- and low-resource languages.