用物品特征生成虚拟用户，提升推荐系统个性化能力
Collaborative Diffusion Model for Recommender System
- 通过物品特征生成伪用户，融合真实与伪用户协同信号
- 在三个数据集上优于现有方法，有效减少个性化信息损失
- 适合需要提升长尾物品推荐效果的场景
基于扩散的推荐系统（DR）因其强大的生成和去噪能力受到关注。然而，现有方法存在两大局限：一是通过注入噪声增强生成能力时会损失个性化信息；二是未能充分利用丰富的物品侧信息。为此，我们提出协同扩散推荐模型（CDiff4Rec）。该模型从物品特征生成伪用户，并通过行为相似性识别真实与伪用户的协同邻居，从而有效重构细粒度用户偏好。在三个公开数据集上的实验表明，通过融合物品内容与协同信号，CDiff4Rec显著缓解了个性化信息丢失问题，性能超越现有基线。
原文摘要 · Abstract (English)
Diffusion-based recommender systems (DR) have gained increasing attention for their advanced generative and denoising capabilities. However, existing DR face two central limitations: (i) a trade-off between enhancing generative capacity via noise injection and retaining the loss of personalized information. (ii) the underutilization of rich item-side information. To address these challenges, we present a Collaborative Diffusion model for Recommender System (CDiff4Rec). Specifically, CDiff4Rec generates pseudo-users from item features and leverages collaborative signals from both real and pseudo personalized neighbors identified through behavioral similarity, thereby effectively reconstructing nuanced user preferences. Experimental results on three public datasets show that CDiff4Rec outperforms competitors by effectively mitigating the loss of personalized information through the integration of item content and collaborative signals.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。