{"ID":2843875,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.06937","arxiv_id":"2511.06937","title":"Fine-Tuning Diffusion-Based Recommender Systems via Reinforcement Learning with Reward Function Optimization","abstract":"Diffusion models recently emerged as a powerful paradigm for recommender systems, offering state-of-the-art performance by modeling the generative process of user-item interactions. However, training such models from scratch is both computationally expensive and yields diminishing returns once convergence is reached. To remedy these challenges, we propose ReFiT, a new framework that integrates Reinforcement learning (RL)-based Fine-Tuning into diffusion-based recommender systems. In contrast to prior RL approaches for diffusion models depending on external reward models, ReFiT adopts a task-aligned design: it formulates the denoising trajectory as a Markov decision process (MDP) and incorporates a collaborative signal-aware reward function that directly reflects recommendation quality. By tightly coupling the MDP structure with this reward signal, ReFiT empowers the RL agent to exploit high-order connectivity for fine-grained optimization, while avoiding the noisy or uninformative feedback common in naive reward designs. Leveraging policy gradient optimization, ReFiT maximizes exact log-likelihood of observed interactions, thereby enabling effective post hoc fine-tuning of diffusion recommenders. Comprehensive experiments on wide-ranging real-world datasets demonstrate that the proposed ReFiT framework (a) exhibits substantial performance gains over strong competitors (up to 36.3% on sequential recommendation), (b) demonstrates strong efficiency with linear complexity in the number of users or items, and (c) generalizes well across multiple diffusion-based recommendation scenarios. The source code and datasets are publicly available at https://anonymous.4open.science/r/ReFiT-4C60.","short_abstract":"Diffusion models recently emerged as a powerful paradigm for recommender systems, offering state-of-the-art performance by modeling the generative process of user-item interactions. However, training such models from scratch is both computationally expensive and yields diminishing returns once convergence is reached. T...","url_abs":"https://arxiv.org/abs/2511.06937","url_pdf":"https://arxiv.org/pdf/2511.06937v1","authors":"[\"Yu Hou\",\"Hua Li\",\"Ha Young Kim\",\"Won-Yong Shin\"]","published":"2025-11-10T10:38:16Z","proceeding":"cs.IR","tasks":"[\"cs.IR\",\"cs.AI\",\"cs.LG\",\"cs.NI\",\"cs.SI\"]","methods":"[\"Reinforcement Learning\",\"Diffusion Model\"]","project_urls":"[\"https://anonymous.4open.science/r/ReFiT-4C60\"]","has_code":false}