Argument-Centric Causal Intervention Method for Mitigating Bias in Cross-Document Event Coreference Resolution

TL;DR

This paper introduces a causal intervention approach for cross-document event coreference resolution that reduces bias from lexical triggers, leading to improved accuracy without additional data augmentation.

Contribution

It proposes a novel Argument-Centric Causal Intervention (ACCI) method that employs causal graphs and counterfactual reasoning to debias coreference models in an end-to-end framework.

Findings

Achieves state-of-the-art F1 scores of 88.4% on ECB+ and 85.2% on GVC datasets.

Effectively reduces spurious lexical correlations in coreference resolution.

Operates without costly data augmentation or heuristic filtering.

Abstract

Cross-document Event Coreference Resolution (CD-ECR) is a fundamental task in natural language processing (NLP) that seeks to determine whether event mentions across multiple documents refer to the same real-world occurrence. However, current CD-ECR approaches predominantly rely on trigger features within input mention pairs, which induce spurious correlations between surface-level lexical features and coreference relationships, impairing the overall performance of the models. To address this issue, we propose a novel cross-document event coreference resolution method based on Argument-Centric Causal Intervention (ACCI). Specifically, we construct a structural causal graph to uncover confounding dependencies between lexical triggers and coreference labels, and introduce backdoor-adjusted interventions to isolate the true causal effect of argument semantics. To further mitigate spurious correlations, ACCI integrates a counterfactual reasoning module that quantifies the causal influence of trigger word perturbations, and an argument-aware enhancement module to promote greater sensitivity to semantically grounded information. In contrast to prior methods that depend on costly data augmentation or heuristic-based filtering, ACCI enables effective debiasing in a unified end-to-end framework without altering the underlying training procedure. Extensive experiments demonstrate that ACCI achieves CoNLL F1 of 88.4% on ECB+ and 85.2% on GVC, achieving state-of-the-art performance. The implementation and materials are available at https://github.com/era211/ACCI.