GRIT: Generative Role-filler Transformers for Document-level Event Entity Extraction
This addresses the problem of extracting entities across document boundaries for template filling, which is incremental as it builds on transformer methods for a specific NLP task.
The authors tackled document-level role-filler entity extraction for template filling by introducing GRIT, a generative transformer-based encoder-decoder framework that models context across sentences and handles coreference and cross-role dependencies, showing substantially better performance than prior work on the MUC-4 dataset.
We revisit the classic problem of document-level role-filler entity extraction (REE) for template filling. We argue that sentence-level approaches are ill-suited to the task and introduce a generative transformer-based encoder-decoder framework (GRIT) that is designed to model context at the document level: it can make extraction decisions across sentence boundaries; is implicitly aware of noun phrase coreference structure, and has the capacity to respect cross-role dependencies in the template structure. We evaluate our approach on the MUC-4 dataset, and show that our model performs substantially better than prior work. We also show that our modeling choices contribute to model performance, e.g., by implicitly capturing linguistic knowledge such as recognizing coreferent entity mentions.