Open Your Eyes: Vision Enhances Message Passing Neural Networks in Link Prediction

Yanbin Wei, Xuehao Wang, Zhan Zhuang, Yang Chen, Shuhao Chen, Yulong Zhang, James Kwok, Yu Zhang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:66259-66279, 2025.

Abstract

Message-passing graph neural networks (MPNNs) and structural features (SFs) are cornerstones for the link prediction task. However, as a common and intuitive mode of understanding, the potential of visual perception has been overlooked in the MPNN community. For the first time, we equip MPNNs with vision structural awareness by proposing an effective framework called Graph Vision Network (GVN), along with a more efficient variant (E-GVN). Extensive empirical results demonstrate that with the proposed frameworks, GVN consistently benefits from the vision enhancement across seven link prediction datasets, including challenging large-scale graphs. Such improvements are compatible with existing state-of-the-art (SOTA) methods and GVNs achieve new SOTA results, thereby underscoring a promising novel direction for link prediction.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-wei25m, title = {Open Your Eyes: Vision Enhances Message Passing Neural Networks in Link Prediction}, author = {Wei, Yanbin and Wang, Xuehao and Zhuang, Zhan and Chen, Yang and Chen, Shuhao and Zhang, Yulong and Kwok, James and Zhang, Yu}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {66259--66279}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/wei25m/wei25m.pdf}, url = {https://proceedings.mlr.press/v267/wei25m.html}, abstract = {Message-passing graph neural networks (MPNNs) and structural features (SFs) are cornerstones for the link prediction task. However, as a common and intuitive mode of understanding, the potential of visual perception has been overlooked in the MPNN community. For the first time, we equip MPNNs with vision structural awareness by proposing an effective framework called Graph Vision Network (GVN), along with a more efficient variant (E-GVN). Extensive empirical results demonstrate that with the proposed frameworks, GVN consistently benefits from the vision enhancement across seven link prediction datasets, including challenging large-scale graphs. Such improvements are compatible with existing state-of-the-art (SOTA) methods and GVNs achieve new SOTA results, thereby underscoring a promising novel direction for link prediction.} }
Endnote
%0 Conference Paper %T Open Your Eyes: Vision Enhances Message Passing Neural Networks in Link Prediction %A Yanbin Wei %A Xuehao Wang %A Zhan Zhuang %A Yang Chen %A Shuhao Chen %A Yulong Zhang %A James Kwok %A Yu Zhang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-wei25m %I PMLR %P 66259--66279 %U https://proceedings.mlr.press/v267/wei25m.html %V 267 %X Message-passing graph neural networks (MPNNs) and structural features (SFs) are cornerstones for the link prediction task. However, as a common and intuitive mode of understanding, the potential of visual perception has been overlooked in the MPNN community. For the first time, we equip MPNNs with vision structural awareness by proposing an effective framework called Graph Vision Network (GVN), along with a more efficient variant (E-GVN). Extensive empirical results demonstrate that with the proposed frameworks, GVN consistently benefits from the vision enhancement across seven link prediction datasets, including challenging large-scale graphs. Such improvements are compatible with existing state-of-the-art (SOTA) methods and GVNs achieve new SOTA results, thereby underscoring a promising novel direction for link prediction.
APA
Wei, Y., Wang, X., Zhuang, Z., Chen, Y., Chen, S., Zhang, Y., Kwok, J. & Zhang, Y.. (2025). Open Your Eyes: Vision Enhances Message Passing Neural Networks in Link Prediction. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:66259-66279 Available from https://proceedings.mlr.press/v267/wei25m.html.

Related Material