Efficient One-Pass End-to-End Entity Linking for Questions
This work addresses entity linking for question answering systems, offering a significant performance boost and fast inference, though it is incremental as it builds on existing biencoder approaches.
The authors tackled the problem of entity linking for questions by introducing ELQ, a fast end-to-end model that jointly performs mention detection and linking in one pass, achieving F1 improvements of +12.7% on WebQSP and +19.6% on GraphQuestions compared to previous state-of-the-art methods.
We present ELQ, a fast end-to-end entity linking model for questions, which uses a biencoder to jointly perform mention detection and linking in one pass. Evaluated on WebQSP and GraphQuestions with extended annotations that cover multiple entities per question, ELQ outperforms the previous state of the art by a large margin of +12.7% and +19.6% F1, respectively. With a very fast inference time (1.57 examples/s on a single CPU), ELQ can be useful for downstream question answering systems. In a proof-of-concept experiment, we demonstrate that using ELQ significantly improves the downstream QA performance of GraphRetriever (arXiv:1911.03868). Code and data available at https://github.com/facebookresearch/BLINK/tree/master/elq