Document-level Event Extraction with Efficient End-to-end Learning of Cross-event Dependencies
This addresses the problem of extracting events and their relationships across entire documents for narrative understanding, representing an incremental improvement in efficiency over existing methods.
The paper tackles document-level event extraction by proposing an end-to-end model using Deep Value Networks to capture cross-event dependencies, achieving comparable performance to CRF-based models on ACE05 with significantly higher computational efficiency.
Fully understanding narratives often requires identifying events in the context of whole documents and modeling the event relations. However, document-level event extraction is a challenging task as it requires the extraction of event and entity coreference, and capturing arguments that span across different sentences. Existing works on event extraction usually confine on extracting events from single sentences, which fail to capture the relationships between the event mentions at the scale of a document, as well as the event arguments that appear in a different sentence than the event trigger. In this paper, we propose an end-to-end model leveraging Deep Value Networks (DVN), a structured prediction algorithm, to efficiently capture cross-event dependencies for document-level event extraction. Experimental results show that our approach achieves comparable performance to CRF-based models on ACE05, while enjoys significantly higher computational efficiency.