{"ID":2838168,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.17888","arxiv_id":"2511.17888","title":"MINDiff: Mask-Integrated Negative Attention for Controlling Overfitting in Text-to-Image Personalization","abstract":"In the personalization process of large-scale text-to-image models, overfitting often occurs when learning specific subject from a limited number of images. Existing methods, such as DreamBooth, mitigate this issue through a class-specific prior-preservation loss, which requires increased computational cost during training and limits user control during inference time. To address these limitations, we propose Mask-Integrated Negative Attention Diffusion (MINDiff). MINDiff introduces a novel concept, negative attention, which suppresses the subject's influence in masked irrelevant regions. We achieve this by modifying the cross-attention mechanism during inference. This enables semantic control and improves text alignment by reducing subject dominance in irrelevant regions. Additionally, during the inference time, users can adjust a scale parameter lambda to balance subject fidelity and text alignment. Our qualitative and quantitative experiments on DreamBooth models demonstrate that MINDiff mitigates overfitting more effectively than class-specific prior-preservation loss. As our method operates entirely at inference time and does not alter the model architecture, it can be directly applied to existing DreamBooth models without re-training. Our code is available at https://github.com/seuleepy/MINDiff.","short_abstract":"In the personalization process of large-scale text-to-image models, overfitting often occurs when learning specific subject from a limited number of images. Existing methods, such as DreamBooth, mitigate this issue through a class-specific prior-preservation loss, which requires increased computational cost during trai...","url_abs":"https://arxiv.org/abs/2511.17888","url_pdf":"https://arxiv.org/pdf/2511.17888v1","authors":"[\"Seulgi Jeong\",\"Jaeil Kim\"]","published":"2025-11-22T02:32:19Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":606740,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2838168,"paper_url":"https://arxiv.org/abs/2511.17888","paper_title":"MINDiff: Mask-Integrated Negative Attention for Controlling Overfitting in Text-to-Image Personalization","repo_url":"https://github.com/seuleepy/MINDiff","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}