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

arXiv: https://arxiv.org/abs/2506.01488 · HTML: https://mlsanity.com/paper/2506.01488/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 59 references that are in the map.
- First posted on arXiv 2 Jun 2025.

## Builds on (most cited first, 21 of 21)

- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Unbiased Scene Graph Generation from Biased Training](https://mlsanity.com/paper/2002.11949/) (2020)
- [Causal Intervention for Leveraging Popularity Bias in Recommendation](https://mlsanity.com/paper/2105.06067/) (2021)
- [Causal Attention for Vision-Language Tasks](https://mlsanity.com/paper/2103.03493/) (2021)
- [Deconfounded Image Captioning: A Causal Retrospect](https://mlsanity.com/paper/2003.03923/) (2020)
- [Clicks can be Cheating: Counterfactual Recommendation for Mitigating Clickbait Issue](https://mlsanity.com/paper/2009.09945/) (2020)
- [Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution](https://mlsanity.com/paper/1906.01753/) (2019)
- [Context De-confounded Emotion Recognition](https://mlsanity.com/paper/2303.11921/) (2023)
- [Cross-document Coreference Resolution over Predicted Mentions](https://mlsanity.com/paper/2106.01210/) (2021)
- [Counterfactually-Augmented SNLI Training Data Does Not Yield Better Generalization Than Unaugmented Data](https://mlsanity.com/paper/2010.04762/) (2020)
- [De-biasing Distantly Supervised Named Entity Recognition via Causal Intervention](https://mlsanity.com/paper/2106.09233/) (2021)
- [Generalizing Cross-Document Event Coreference Resolution Across Multiple Corpora](https://mlsanity.com/paper/2011.12249/) (2020)
- [Focus on what matters: Applying Discourse Coherence Theory to Cross Document Coreference](https://mlsanity.com/paper/2110.05362/) (2021)
- [What happens before and after: Multi-Event Commonsense in Event Coreference Resolution](https://mlsanity.com/paper/2302.09715/) (2023)
- [End-to-End Neural Event Coreference Resolution](https://mlsanity.com/paper/2009.08153/) (2020)
- [Pairwise Representation Learning for Event Coreference](https://mlsanity.com/paper/2010.12808/) (2020)
- [Linear Cross-document Event Coreference Resolution with X-AMR](https://mlsanity.com/paper/2404.08656/) (2024)
- [A Rationale-centric Counterfactual Data Augmentation Method for Cross-Document Event Coreference Resolution](https://mlsanity.com/paper/2404.01921/) (2024)
- [Enhancing Cross-Document Event Coreference Resolution by Discourse Structure and Semantic Information](https://mlsanity.com/paper/2406.15990/) (2024)
- [CorefPrompt: Prompt-based Event Coreference Resolution by Measuring Event Type and Argument Compatibilities](https://mlsanity.com/paper/2310.14512/) (2023)
- [Okay, Let's Do This! Modeling Event Coreference with Generated Rationales and Knowledge Distillation](https://mlsanity.com/paper/2404.03196/) (2024)

## Abstract (arXiv)

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.