Unleashing the Potential of Neighbors: Diffusion-based Latent Neighbor Generation for Session-based Recommendation

2026-01-07

arXiv.org

AI Summary

This work addresses the limitation of existing session-based recommendation methods that rely solely on explicitly adjacent sessions, which struggle to alleviate data sparsity. To overcome this, we propose DiffSBR, a novel model that introduces a diffusion mechanism to generate high-quality implicit neighbors in the interest space. Specifically, a retrieval-augmented diffusion module identifies potentially relevant sessions, while a self-augmented diffusion component refines the generation process. Furthermore, multimodal signals and contrastive learning are integrated to enhance session representations. By moving beyond dependence on explicit neighbors, DiffSBR achieves significant performance gains over state-of-the-art methods across four public benchmarks, demonstrating the effectiveness of the proposed implicit neighbor generation strategy and its positive impact on recommendation accuracy.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Multimodal Learning
• Computer Vision: Diffusion Models for Vision

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Session-based recommendation aims to predict the next item that anonymous users may be interested in, based on their current session interactions. Recent studies have demonstrated that retrieving neighbor sessions to augment the current session can effectively alleviate the data sparsity issue and improve recommendation performance. However, existing methods typically rely on explicitly observed session data, neglecting latent neighbors - not directly observed but potentially relevant within the interest space - thereby failing to fully exploit the potential of neighbor sessions in recommendation. To address the above limitation, we propose a novel model of diffusion-based latent neighbor generation for session-based recommendation, named DiffSBR. Specifically, DiffSBR leverages two diffusion modules, including retrieval-augmented diffusion and self-augmented diffusion, to generate high-quality latent neighbors. In the retrieval-augmented diffusion module, we leverage retrieved neighbors as guiding signals to constrain and reconstruct the distribution of latent neighbors. Meanwhile, we adopt a training strategy that enables the retriever to learn from the feedback provided by the generator. In the self-augmented diffusion module, we explicitly guide the generation of latent neighbors by injecting the current session's multi-modal signals through contrastive learning. After obtaining the generated latent neighbors, we utilize them to enhance session representations for improving session-based recommendation. Extensive experiments on four public datasets show that DiffSBR generates effective latent neighbors and improves recommendation performance against state-of-the-art baselines.

Problem

• session-based recommendation
• latent neighbors
• data sparsity
• neighbor sessions
• interest space

Innovation

• diffusion model
• latent neighbor generation
• session-based recommendation
• retrieval-augmented diffusion
• contrastive learning