SpEL: Structured Prediction for Entity Linking
Submission Type: Regular Long Paper
Submission Track: Information Extraction
Submission Track 2: Efficient Methods for NLP
Keywords: Structured Prediction, Entity Linking, AIDA dataset, AIDA/testc, SpEL
TL;DR: SpEL is a structured prediction entity linking approach that uses new training and inference ideas obtaining a new state of the art on Wikipedia entity linking, with better compute efficiency and faster inference than previous methods.
Abstract: Entity linking is a prominent thread of research focused on structured data creation by linking spans of text to an ontology or knowledge source. We revisit the use of structured prediction for entity linking which classifies each individual input token as an entity, and aggregates the token predictions. Our system, called SpEL (Structured prediction for Entity Linking) is a state-of-the-art entity linking system that uses some new ideas to apply structured prediction to the task of entity linking including: two refined fine-tuning steps; a context sensitive prediction aggregation strategy; reduction of the size of the model's output vocabulary, and; we address a common problem in entity-linking systems where there is a training vs. inference tokenization mismatch. Our experiments show that we can outperform the state-of-the-art on the commonly used AIDA benchmark dataset for entity linking to Wikipedia. Our method is also very compute efficient in terms of number of parameters and speed of inference.
Submission Number: 4416
Loading