Dynamic Prefix-Tuning for Generative Template-based Event Extraction

TL;DR

This paper introduces GTEE-DynPref, a dynamic prefix-tuning method for generative event extraction that adapts to context and improves performance over static approaches, excelling on ACE 2005 and ERE datasets.

Contribution

It proposes a novel dynamic prefix-tuning approach for generative event extraction that enhances adaptability and performance over existing static prompt methods.

Findings

Achieves state-of-the-art results on ERE dataset.

Performs competitively with OneIE on ACE 2005.

Effectively adapts to new event types.

Abstract

We consider event extraction in a generative manner with template-based conditional generation. Although there is a rising trend of casting the task of event extraction as a sequence generation problem with prompts, these generation-based methods have two significant challenges, including using suboptimal prompts and static event type information. In this paper, we propose a generative template-based event extraction method with dynamic prefix (GTEE-DynPref) by integrating context information with type-specific prefixes to learn a context-specific prefix for each context. Experimental results show that our model achieves competitive results with the state-of-the-art classification-based model OneIE on ACE 2005 and achieves the best performances on ERE. Additionally, our model is proven to be portable to new types of events effectively.