Yuzhe Chen

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.

Multi-Behavior Recommendation (MBR) aims to model personalized user preferences by integrating diverse interaction behaviors (e.g., page view, favorite, add to cart, purchase). However, latent confounders such as contextual influences and social relationships can obscure the true causal effects in real-world scenarios, thereby confounding the model’s prediction. Although existing MBR research extensively explores behavioral dependencies and heterogeneity, it frequently overlooks the impact of latent confounders, thereby limiting its ability to capture users’ genuine preferences. To address the limitations of existing methods, we identify two key challenges in MBR: (1) how to infer latent confounders, and (2) how to mitigate their influence across multi-behavior interactions. To this end, we propose Causal Variational Inference for Deconfounded (CVID) MBR. CVID employs a variational graph autoencoder to model latent uncertainty in multi-behavior interactions and introduces a confounder inference module to generate behavior-specific latent confounders via variational inference. In the conditional diffusion module, noise is progressively injected during the forward process to simulate the dynamic evolution of user preferences, while the reverse process leverages the inferred latent confounders to guide denoising through back-door adjustment, thereby recovering the true causal effects between multi-behavior interactions and the model’s prediction. Extensive experiments on public multi-behavior datasets demonstrate that CVID consistently outperforms state-of-the-art baselines in mitigating confounding effects and improving recommendation accuracy, validating its effectiveness and superiority.

Sequential recommendation (SR) predicts the user’s future preferences based on the sequence of interactions. Recently, some methods for SR have utilized contrastive learning to incorporate self-supervised signals into SR to alleviate the data sparsity problem. Despite these achievements, they overlook the fact that users’ multi-behavior interactions in real-world scenarios (e.g., page view, favorite, add to cart, and purchase). Moreover, they disregard the temporal dependencies in users’ preferences and their influence on attribute information, leading to models that struggle to accurately capture users’ personalized preferences. Therefore, we propose a multi-behavior collaborative contrastive learning for sequential recommendation model. First, we introduce both user-side and item-side attribute information and design an attribute-weight-enhanced attention in multi-behavior interaction scenarios. It enhances the model’s ability to capture user’s multi-behavior preferences while considering the influence of attribute information. Second, in order to capture users’ fine-grained temporal preferences. We divide the interaction sequences into different time scales based on the users’ multi-behavior interaction timestamps. In addition, introduce temporal aware attention to generate temporal embeddings for different time scales and effectively integrate them with the user’s multi-behavior embeddings. Finally, we design collaborative contrastive learning, which collaboratively captures users’ multi-behavior personalized preferences from both temporal and attribute perspectives. This approach alleviates the issue of data sparsity. We conduct extensive experiments on two datasets to validate the effectiveness and superiority of our model.