Beyond Verbal Cues: Emotional Contagion Graph Network for Causal Emotion Entailment

In this paper, we introduce the Emotional Contagion Graph Network (ECGN), a novel model designed to improve causal emotion entailment in conversations by simulating both explicit (verbal) and implicit (non-verbal) emotional influences.

More details can be found in our paper: Fangxu Yu, Junjie Guo, Zhen Wu, Xinyu Dai, "Beyond Verbal Cues: Emotional Contagion Graph Network for Causal Emotion Entailment"

ECGN

As illustrated in the above diagram, our ECGN framework includes four steps:

1. Context Encoding: Use pretrained language models to extract both the utterance and emotions to obtain the encodings of them.
2. Graph Construction: Construct a heterogeneous graph modeling the complex interaction relations, including the simulated implicit and explicit emotional contagion.
3. Emotional Dynamic Interaction: Build up a graph-learning model for learning dynamics between different node features.
4. Cause Prediction: Use a cause prediction module to identify the causes of emotions within the conversation.

Code

1) Download this GitHub

git clone https://github.com/Yu-Fangxu/ECGN.git

3) Run Command for ECGN

bash run.sh 

Citation

@inproceedings{yu2025beyond,
  title={Beyond Verbal Cues: Emotional Contagion Graph Network for Causal Emotion Entailment},
  author={Yu, Fangxu and Guo, Junjie and Wu, Zhen and Dai, Xinyu},
  booktitle={Findings of the Association for Computational Linguistics: ACL 2025},
  pages={1755--1767},
  year={2025}
}