Flow Matching for Denoised Social Recommendation

Yinxuan Huang, Ke Liang, Zhuofan Dong, Xiaodong Qu, Wang Tianxiang, Yue Han, Jingao Xu, Bin Zhou, Ye Wang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:25474-25485, 2025.

Abstract

Graph-based social recommendation (SR) models suffer from various noises of the social graphs, hindering their recommendation performances. Either graph-level redundancy or graph-level missing will indeed influence the social graph structures, further influencing the message propagation procedure of graph neural networks (GNNs). Generative models, especially diffusion-based models, are usually used to reconstruct and recover the data in better quality from original data with noises. Motivated by it, a few works take attempts on it for social recommendation. However, they can only handle isotropic Gaussian noises but fail to leverage the anisotropic ones. Meanwhile the anisotropic relational structures in social graphs are commonly seen, so that existing models cannot sufficiently utilize the graph structures, which constraints the capacity of noise removal and recommendation performances. Compared to the diffusion strategy, the flow matching strategy shows better ability to handle the data with anisotropic noises since they can better preserve the data structures during the learning procedure. Inspired by it, we propose RecFlow which is the first flow-matching based SR model. Concretely, RecFlow performs flow matching on the structure representations of social graphs. Then, a conditional learning procedure is designed for optimization. Extensive performances prove the promising performances of our RecFlow from six aspects, including superiority, effectiveness, robustnesses, sensitivity, convergence and visualization.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-huang25p, title = {Flow Matching for Denoised Social Recommendation}, author = {Huang, Yinxuan and Liang, Ke and Dong, Zhuofan and Qu, Xiaodong and Tianxiang, Wang and Han, Yue and Xu, Jingao and Zhou, Bin and Wang, Ye}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {25474--25485}, 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/huang25p/huang25p.pdf}, url = {https://proceedings.mlr.press/v267/huang25p.html}, abstract = {Graph-based social recommendation (SR) models suffer from various noises of the social graphs, hindering their recommendation performances. Either graph-level redundancy or graph-level missing will indeed influence the social graph structures, further influencing the message propagation procedure of graph neural networks (GNNs). Generative models, especially diffusion-based models, are usually used to reconstruct and recover the data in better quality from original data with noises. Motivated by it, a few works take attempts on it for social recommendation. However, they can only handle isotropic Gaussian noises but fail to leverage the anisotropic ones. Meanwhile the anisotropic relational structures in social graphs are commonly seen, so that existing models cannot sufficiently utilize the graph structures, which constraints the capacity of noise removal and recommendation performances. Compared to the diffusion strategy, the flow matching strategy shows better ability to handle the data with anisotropic noises since they can better preserve the data structures during the learning procedure. Inspired by it, we propose RecFlow which is the first flow-matching based SR model. Concretely, RecFlow performs flow matching on the structure representations of social graphs. Then, a conditional learning procedure is designed for optimization. Extensive performances prove the promising performances of our RecFlow from six aspects, including superiority, effectiveness, robustnesses, sensitivity, convergence and visualization.} }
Endnote
%0 Conference Paper %T Flow Matching for Denoised Social Recommendation %A Yinxuan Huang %A Ke Liang %A Zhuofan Dong %A Xiaodong Qu %A Wang Tianxiang %A Yue Han %A Jingao Xu %A Bin Zhou %A Ye Wang %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-huang25p %I PMLR %P 25474--25485 %U https://proceedings.mlr.press/v267/huang25p.html %V 267 %X Graph-based social recommendation (SR) models suffer from various noises of the social graphs, hindering their recommendation performances. Either graph-level redundancy or graph-level missing will indeed influence the social graph structures, further influencing the message propagation procedure of graph neural networks (GNNs). Generative models, especially diffusion-based models, are usually used to reconstruct and recover the data in better quality from original data with noises. Motivated by it, a few works take attempts on it for social recommendation. However, they can only handle isotropic Gaussian noises but fail to leverage the anisotropic ones. Meanwhile the anisotropic relational structures in social graphs are commonly seen, so that existing models cannot sufficiently utilize the graph structures, which constraints the capacity of noise removal and recommendation performances. Compared to the diffusion strategy, the flow matching strategy shows better ability to handle the data with anisotropic noises since they can better preserve the data structures during the learning procedure. Inspired by it, we propose RecFlow which is the first flow-matching based SR model. Concretely, RecFlow performs flow matching on the structure representations of social graphs. Then, a conditional learning procedure is designed for optimization. Extensive performances prove the promising performances of our RecFlow from six aspects, including superiority, effectiveness, robustnesses, sensitivity, convergence and visualization.
APA
Huang, Y., Liang, K., Dong, Z., Qu, X., Tianxiang, W., Han, Y., Xu, J., Zhou, B. & Wang, Y.. (2025). Flow Matching for Denoised Social Recommendation. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:25474-25485 Available from https://proceedings.mlr.press/v267/huang25p.html.

Related Material