Explaining Graph Neural Networks with Large Language Models: A Counterfactual Perspective on Molecule Graphs

Yinhan He, Zaiyi Zheng, Patrick Soga, Yaochen Zhu, Yushun Dong, Jundong Li


Abstract
In recent years, Graph Neural Networks (GNNs) have become successful in molecular property prediction tasks such as toxicity analysis. However, due to the black-box nature of GNNs, their outputs can be concerning in high-stakes decision-making scenarios, e.g., drug discovery. Facing such an issue, Graph Counterfactual Explanation (GCE) has emerged as a promising approach to improve GNN transparency. However, current GCE methods usually fail to take domain-specific knowledge into consideration, which can result in outputs that are not easily comprehensible by humans. To address this challenge, we propose a novel GCE method, LLM-GCE, to unleash the power of large language models (LLMs) in explaining GNNs for molecular property prediction. Specifically, we utilize an autoencoder to generate the counterfactual graph topology from a set of counterfactual text pairs (CTPs) based on an input graph. Meanwhile, we also incorporate a CTP dynamic feedback module to mitigate LLM hallucination, which provides intermediate feedback derived from the generated counterfactuals as an attempt to give more faithful guidance. Extensive experiments demonstrate the superior performance of LLM-GCE. Our code is released on https://github.com/YinhanHe123/new_LLM4GNNExplanation.
Anthology ID:
2024.findings-emnlp.415
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2024
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7079–7096
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.415/
DOI:
10.18653/v1/2024.findings-emnlp.415
Bibkey:
Cite (ACL):
Yinhan He, Zaiyi Zheng, Patrick Soga, Yaochen Zhu, Yushun Dong, and Jundong Li. 2024. Explaining Graph Neural Networks with Large Language Models: A Counterfactual Perspective on Molecule Graphs. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 7079–7096, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Explaining Graph Neural Networks with Large Language Models: A Counterfactual Perspective on Molecule Graphs (He et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.415.pdf