TY - JOUR

T1 - Diffusion Contrastive Learning Model for Multi-Behavior Recommendation

AU - Chen, Yuzhe

AU - Cao, Jie

AU - Wang, Youquan

AU - Wu, Jia

AU - Xu, Guandong

AU - Chen, Huanhuan

AU - Вукович, Дарко

N1 - Chen, Y., Cao, J., Wang, Y., Wu, J., Xu, G., Chen, H., ... & Vukovic, D. B. (2026). Diffusion Contrastive Learning Model for Multi-Behavior Recommendation. IEEE Transactions on Big Data.

PY - 2026

Y1 - 2026

N2 - Multi-behavior recommendation leverages diverse user interactions to capture personalized user preferences. To alleviate the data sparsity of the crucial target behavior, multi-behavior contrastive learning (MBCL) improves performance by enriching self-supervised signals from auxiliary behaviors. However, existing methods rely on heuristic augmentation techniques to construct contrastive views, which disrupt essential multi-behavior collaborative information; simultaneously, auxiliary behaviors, influenced by multiple confounders, easily introduce semantic confusion into the MBCL alignment process. Inspired by recent advances of diffusion models, we propose DiffMB, a novel diffusion contrastive learning framework. DiffMB first employs a Multi-Behavior Weight-Enhanced Graph Neural Network to capture behavioral heterogeneity through parallel encoding. At the core of DiffMB is a Hybrid Contrastive Learning (HCL) strategy that tackles this dilemma via two synergistic components. First, Target-Behavior Diffusion Contrastive Learning leverages a diffusion process to generate high-quality, semantically consistent views for the target behavior, thereby consolidating its semantics while avoiding the disruption of collaborative information. Second, Multi-Behavior Collaborative Contrastive Learning extracts valuable collaborative signals from auxiliary behaviors. Furthermore, a Dynamic Optimization Strategy is introduced to reconcile gradient conflicts in multi-task training. Extensive experiments on three datasets demonstrate that DiffMB significantly outperforms state-of-the-art baselines.

AB - Multi-behavior recommendation leverages diverse user interactions to capture personalized user preferences. To alleviate the data sparsity of the crucial target behavior, multi-behavior contrastive learning (MBCL) improves performance by enriching self-supervised signals from auxiliary behaviors. However, existing methods rely on heuristic augmentation techniques to construct contrastive views, which disrupt essential multi-behavior collaborative information; simultaneously, auxiliary behaviors, influenced by multiple confounders, easily introduce semantic confusion into the MBCL alignment process. Inspired by recent advances of diffusion models, we propose DiffMB, a novel diffusion contrastive learning framework. DiffMB first employs a Multi-Behavior Weight-Enhanced Graph Neural Network to capture behavioral heterogeneity through parallel encoding. At the core of DiffMB is a Hybrid Contrastive Learning (HCL) strategy that tackles this dilemma via two synergistic components. First, Target-Behavior Diffusion Contrastive Learning leverages a diffusion process to generate high-quality, semantically consistent views for the target behavior, thereby consolidating its semantics while avoiding the disruption of collaborative information. Second, Multi-Behavior Collaborative Contrastive Learning extracts valuable collaborative signals from auxiliary behaviors. Furthermore, a Dynamic Optimization Strategy is introduced to reconcile gradient conflicts in multi-task training. Extensive experiments on three datasets demonstrate that DiffMB significantly outperforms state-of-the-art baselines.

KW - Contrastive learning

KW - Diffusion model

KW - Graph neural network

KW - Multi-behavior recommendation

UR - https://ieeexplore.ieee.org/abstract/document/11415640/authors#authors

UR - https://www.scopus.com/pages/publications/105031660466

U2 - 10.1109/tbdata.2026.3668649

DO - 10.1109/tbdata.2026.3668649

M3 - Article

JO - IEEE Transactions on Big Data

JF - IEEE Transactions on Big Data

SN - 2332-7790

ER -