Efficient Document-level Event Extraction via Pseudo-Trigger-aware Pruned Complete Graph

TL;DR

This paper introduces PTPCG, a fast, lightweight, non-autoregressive model for document-level event extraction that uses pseudo triggers and pruned complete graphs to improve efficiency and maintain competitive accuracy.

Contribution

The paper presents a novel non-autoregressive decoding method with pseudo trigger guidance and graph pruning, significantly reducing resource consumption and inference time in event extraction.

Findings

Achieves 19.8% fewer parameters with competitive results.

Only 3.8% GPU hours needed for training.

Up to 8.5 times faster inference than previous methods.

Abstract

Most previous studies of document-level event extraction mainly focus on building argument chains in an autoregressive way, which achieves a certain success but is inefficient in both training and inference. In contrast to the previous studies, we propose a fast and lightweight model named as PTPCG. In our model, we design a novel strategy for event argument combination together with a non-autoregressive decoding algorithm via pruned complete graphs, which are constructed under the guidance of the automatically selected pseudo triggers. Compared to the previous systems, our system achieves competitive results with 19.8\% of parameters and much lower resource consumption, taking only 3.8\% GPU hours for training and up to 8.5 times faster for inference. Besides, our model shows superior compatibility for the datasets with (or without) triggers and the pseudo triggers can be the supplements for annotated triggers to make further improvements. Codes are available at https://github.com/Spico197/DocEE .