Dynamic Global Memory for Document-level Argument Extraction

TL;DR

This paper introduces a global memory framework for document-level event argument extraction that captures comprehensive context and improves robustness, outperforming prior methods.

Contribution

The paper proposes a novel global neural generation framework with a document memory store for enhanced document-level argument extraction.

Findings

Outperforms prior methods significantly.

More robust to adversarial annotations.

Effective in capturing global document context.

Abstract

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.)