Dynamic Entity Representations in Neural Language Models (1708.00781v1)

Abstract: Understanding a long document requires tracking how entities are introduced and evolve over time. We present a new type of LLM, EntityNLM, that can explicitly model entities, dynamically update their representations, and contextually generate their mentions. Our model is generative and flexible; it can model an arbitrary number of entities in context while generating each entity mention at an arbitrary length. In addition, it can be used for several different tasks such as LLMing, coreference resolution, and entity prediction. Experimental results with all these tasks demonstrate that our model consistently outperforms strong baselines and prior work.