Interests Burn-down Diffusion Process for Personalized Collaborative Filtering

2026-05-06

AI Summary

This work addresses the limitations of conventional diffusion-based recommender models, which rely on Gaussian noise and struggle to capture fine-grained user-item interaction patterns, thereby constraining recommendation performance. To overcome this, the authors propose the Interest Burn-down Process—an innovative non-Gaussian diffusion mechanism that explicitly models the dynamic decay of user interest in candidate items and its reverse restoration during generation. Building upon this mechanism, they introduce StageCF, a framework that integrates interest evolution with reverse generative dynamics to synthesize high-quality personalized interaction data. Extensive experiments demonstrate that the proposed approach significantly outperforms state-of-the-art generative and diffusion-based recommendation models across multiple benchmark datasets, confirming the effectiveness and superiority of the devised interest-aware diffusion paradigm.

Abstract

Generative methods have gained widespread attention in Collaborative Filtering (CF) tasks for their ability to produce high-quality personalized samples aligned with users' interests. Among them, diffusion generative models have raised increasing attention in recommendation field. Despite that the pioneering efforts have applied the conventional diffusion process to model diffusive user interests, the incongruity between the Gaussian noise and the subtle nature of user's personalized interaction behavior has led to sub-optimal results. To this end, we introduce a specifically-tailored diffusion scheme for interaction systems, namely the interests burn-down process. The interests burn-down process delineates the decay of user interests towards candidate items, complemented by its reverse burn-up process that yields personalized recommendation for users. The inherent burn-down nature of this process adeptly models the diffusive user interests, aligning seamlessly with the requirements of CF tasks. We present a novel recommendation method StageCF to illustrate the superiority of this newly proposed diffusion process. Experimental results have demonstrated the effectiveness of StageCF against existing generative and diffusion-based baseline methods. Furthermore, comprehensive studies validate the functionality of interests burn-down process, shedding light on its capacity to generate personalized interactions.

Problem

• Collaborative Filtering
• Diffusion Models
• Personalized Recommendation
• User Interests
• Generative Methods

Innovation

• interests burn-down diffusion
• personalized collaborative filtering
• diffusion generative models
• user interest modeling
• StageCF