{"ID":2866034,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.21167","arxiv_id":"2509.21167","title":"A Unified Framework for Diffusion Model Unlearning with f-Divergence","abstract":"Most existing methods for concept unlearning in text-to-image diffusion models minimize a mean squared error (MSE) loss between the denoiser outputs conditioned on a target and an anchor concept, which is implicitly the KL divergence between two Gaussians. We generalize this objective to any $f$-divergence, recovering MSE as the KL instance, and identify a family of $α$-divergences whose Gaussian closed-form yields cheap, MSE-like training objectives. For the remaining $f$-divergences, we provide a min-max objective based on the variational formulation of the $f$-divergence. We theoretically analyze and numerically validate how different $f$-divergences impact the gradient magnitude and the convergence properties of the algorithm, affecting the quality of unlearning. For instance, we observe that the Hellinger closed-form instance consistently dominates MSE across multiple scenarios. More generally, the proposed unified framework offers a flexible paradigm for selecting the optimal divergence based on the application and user goal, allowing for finer control over the trade-off between unlearning efficacy and generative fidelity.","short_abstract":"Most existing methods for concept unlearning in text-to-image diffusion models minimize a mean squared error (MSE) loss between the denoiser outputs conditioned on a target and an anchor concept, which is implicitly the KL divergence between two Gaussians. We generalize this objective to any $f$-divergence, recovering...","url_abs":"https://arxiv.org/abs/2509.21167","url_pdf":"https://arxiv.org/pdf/2509.21167v2","authors":"[\"Nicola Novello\",\"Federico Fontana\",\"Luigi Cinque\",\"Deniz Gunduz\",\"Andrea M. Tonello\"]","published":"2025-09-25T13:51:04Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false}