Dynamic Global Memory for Document-level Argument Extraction
This addresses the problem of extracting event arguments from news articles for information extraction tasks, representing an incremental improvement over existing methods by better handling global context.
The paper tackles document-level event argument extraction by introducing a global neural generation-based framework with a document memory store to capture contextual event information, which substantially outperforms prior methods and shows robustness to adversarial annotations.
Extracting informative arguments of events from news articles is a challenging problem in information extraction, which requires a global contextual understanding of each document. While recent work on document-level extraction has gone beyond single-sentence and increased the cross-sentence inference capability of end-to-end models, they are still restricted by certain input sequence length constraints and usually ignore the global context between events. To tackle this issue, we introduce a new global neural generation-based framework for document-level event argument extraction by constructing a document memory store to record the contextual event information and leveraging it to implicitly and explicitly help with decoding of arguments for later events. Empirical results show that our framework outperforms prior methods substantially and it is more robust to adversarially annotated examples with our constrained decoding design. (Our code and resources are available at https://github.com/xinyadu/memory_docie for research purpose.)