Linear Cross-document Event Coreference Resolution with X-AMR

TL;DR

This paper introduces X-AMR, a graphical event representation and multi-hop algorithm that simplifies cross-document event coreference resolution, making it more cost-effective, interpretable, and easier to annotate, while evaluating LLMs like GPT-4.

Contribution

It proposes X-AMR and a novel multi-hop algorithm for efficient, interpretable cross-document event coreference resolution, and provides a new annotated benchmark dataset.

Findings

X-AMR simplifies ECR, reducing LLM costs.

The multi-hop algorithm improves coreference accuracy.

GPT-4's performance is comparable to humans with limitations.

Abstract

Event Coreference Resolution (ECR) as a pairwise mention classification task is expensive both for automated systems and manual annotations. The task's quadratic difficulty is exacerbated when using Large Language Models (LLMs), making prompt engineering for ECR prohibitively costly. In this work, we propose a graphical representation of events, X-AMR, anchored around individual mentions using a \textbf{cross}-document version of \textbf{A}bstract \textbf{M}eaning \textbf{R}epresentation. We then linearize the ECR with a novel multi-hop coreference algorithm over the event graphs. The event graphs simplify ECR, making it a) LLM cost-effective, b) compositional and interpretable, and c) easily annotated. For a fair assessment, we first enrich an existing ECR benchmark dataset with these event graphs using an annotator-friendly tool we introduce. Then, we employ GPT-4, the newest LLM by OpenAI, for these annotations. Finally, using the ECR algorithm, we assess GPT-4 against humans and analyze its limitations. Through this research, we aim to advance the state-of-the-art for efficient ECR and shed light on the potential shortcomings of current LLMs at this task. Code and annotations: \url{https://github.com/ahmeshaf/gpt_coref}