An AMR-based Link Prediction Approach for Document-level Event Argument Extraction

TL;DR

This paper introduces a novel link prediction approach using a tailored AMR graph for document-level event argument extraction, significantly improving accuracy and efficiency over previous methods.

Contribution

It reformulates event argument extraction as a link prediction task on a new tailored AMR graph, integrating span and surrounding event information for better performance.

Findings

Outperforms state-of-the-art models by 3.63 F1 points on WikiEvents.

Achieves 2.33 F1 points improvement on RAMS dataset.

Reduces inference time by 56% compared to previous approaches.

Abstract

Recent works have introduced Abstract Meaning Representation (AMR) for Document-level Event Argument Extraction (Doc-level EAE), since AMR provides a useful interpretation of complex semantic structures and helps to capture long-distance dependency. However, in these works AMR is used only implicitly, for instance, as additional features or training signals. Motivated by the fact that all event structures can be inferred from AMR, this work reformulates EAE as a link prediction problem on AMR graphs. Since AMR is a generic structure and does not perfectly suit EAE, we propose a novel graph structure, Tailored AMR Graph (TAG), which compresses less informative subgraphs and edge types, integrates span information, and highlights surrounding events in the same document. With TAG, we further propose a novel method using graph neural networks as a link prediction model to find event arguments. Our extensive experiments on WikiEvents and RAMS show that this simpler approach outperforms the state-of-the-art models by 3.63pt and 2.33pt F1, respectively, and do so with reduced 56% inference time. The code is availabel at https://github.com/ayyyq/TARA.