Zero-Shot Cross-Lingual Document-Level Event Causality Identification with Heterogeneous Graph Contrastive Transfer Learning

TL;DR

This paper introduces a novel zero-shot cross-lingual document-level event causality identification model that leverages heterogeneous graph interactions and contrastive transfer learning, significantly outperforming previous methods and even surpassing GPT-3.5 in multilingual scenarios.

Contribution

It proposes a heterogeneous graph interaction network combined with multi-granularity contrastive transfer learning for improved cross-lingual transfer in document-level ECI.

Findings

Outperforms previous state-of-the-art by 9.4% and 8.2% in monolingual and multilingual F1 scores.

Zero-shot framework exceeds GPT-3.5 with few-shot learning by 24.3%.

Effective modeling of long-distance dependencies and cross-lingual causal knowledge transfer.

Abstract

Event Causality Identification (ECI) refers to the detection of causal relations between events in texts. However, most existing studies focus on sentence-level ECI with high-resource languages, leaving more challenging document-level ECI (DECI) with low-resource languages under-explored. In this paper, we propose a Heterogeneous Graph Interaction Model with Multi-granularity Contrastive Transfer Learning (GIMC) for zero-shot cross-lingual document-level ECI. Specifically, we introduce a heterogeneous graph interaction network to model the long-distance dependencies between events that are scattered over a document. Then, to improve cross-lingual transferability of causal knowledge learned from the source language, we propose a multi-granularity contrastive transfer learning module to align the causal representations across languages. Extensive experiments show our framework outperforms the previous state-of-the-art model by 9.4% and 8.2% of average F1 score on monolingual and multilingual scenarios respectively. Notably, in the multilingual scenario, our zero-shot framework even exceeds GPT-3.5 with few-shot learning by 24.3% in overall performance.