EDIN: An End-to-end Benchmark and Pipeline for Unknown Entity Discovery and Indexing
This addresses the challenge of linking novel or unlisted entities in knowledge bases for natural language processing applications, representing an incremental improvement over existing entity linking methods.
The paper tackles the problem of incomplete knowledge bases in entity linking by introducing the EDIN benchmark for unknown entity discovery and indexing, and shows that indexing a single embedding per entity from multiple mentions outperforms independent mention indexing.
Existing work on Entity Linking mostly assumes that the reference knowledge base is complete, and therefore all mentions can be linked. In practice this is hardly ever the case, as knowledge bases are incomplete and because novel concepts arise constantly. This paper created the Unknown Entity Discovery and Indexing (EDIN) benchmark where unknown entities, that is entities without a description in the knowledge base and labeled mentions, have to be integrated into an existing entity linking system. By contrasting EDIN with zero-shot entity linking, we provide insight on the additional challenges it poses. Building on dense-retrieval based entity linking, we introduce the end-to-end EDIN pipeline that detects, clusters, and indexes mentions of unknown entities in context. Experiments show that indexing a single embedding per entity unifying the information of multiple mentions works better than indexing mentions independently.