Towards A Tri-View Diffusion Framework for Recommendation

2025-11-25

AI Summary

Existing diffusion-based recommendation models lack thermodynamic theoretical foundations, and their energy optimization objectives fundamentally conflict with the conventional entropy minimization paradigm. Method: This paper establishes, for the first time, a thermodynamically grounded framework for recommendation completeness analysis. We propose FreeRec—a minimalist diffusion recommendation model—that unifies energy and entropy optimization via Helmholtz free energy maximization. To preserve the bipartite structure of user–item interactions, we design an anisotropic-preserving denoiser; additionally, we introduce the AR-GSP sampling strategy to enhance generative robustness. Contribution/Results: Theoretical analysis guarantees convergence and preference fidelity. Empirical evaluation demonstrates that FreeRec significantly outperforms state-of-the-art baselines in recommendation accuracy, inference efficiency, and hard negative sample learning.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Diffusion models (DMs) have recently gained significant interest for their exceptional potential in recommendation tasks. This stems primarily from their prominent capability in distilling, modeling, and generating comprehensive user preferences. However, previous work fails to examine DMs in recommendation tasks through a rigorous lens. In this paper, we first experimentally investigate the completeness of recommender models from a thermodynamic view. We reveal that existing DM-based recommender models operate by maximizing the energy, while classic recommender models operate by reducing the entropy. Based on this finding, we propose a minimalistic diffusion framework that incorporates both factors via the maximization of Helmholtz free energy. Meanwhile, to foster the optimization, our reverse process is armed with a well-designed denoiser to maintain the inherent anisotropy, which measures the user-item cross-correlation in the context of bipartite graphs. Finally, we adopt an Acceptance-Rejection Gumbel Sampling Process (AR-GSP) to prioritize the far-outnumbered unobserved interactions for model robustness. AR-GSP integrates an acceptance-rejection sampling to ensure high-quality hard negative samples for general recommendation tasks, and a timestep-dependent Gumbel Softmax to handle an adaptive sampling strategy for diffusion models. Theoretical analyses and extensive experiments demonstrate that our proposed framework has distinct superiority over baselines in terms of accuracy and efficiency.

Problem

• Investigating diffusion models' thermodynamic completeness in recommendation systems
• Proposing a minimalist diffusion framework maximizing Helmholtz free energy
• Developing sampling methods to handle unobserved interactions for robustness

Innovation

• Maximizes Helmholtz free energy for recommendations
• Uses anisotropic denoiser in reverse diffusion process
• Implements acceptance-rejection Gumbel sampling for robustness