{"ID":2888432,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.23652","arxiv_id":"2507.23652","title":"Adaptively Distilled ControlNet: Accelerated Training and Superior Sampling for Medical Image Synthesis","abstract":"Medical image annotation is constrained by privacy concerns and labor-intensive labeling, significantly limiting the performance and generalization of segmentation models. While mask-controllable diffusion models excel in synthesis, they struggle with precise lesion-mask alignment. We propose \\textbf{Adaptively Distilled ControlNet}, a task-agnostic framework that accelerates training and optimization through dual-model distillation. Specifically, during training, a teacher model, conditioned on mask-image pairs, regularizes a mask-only student model via predicted noise alignment in parameter space, further enhanced by adaptive regularization based on lesion-background ratios. During sampling, only the student model is used, enabling privacy-preserving medical image generation. Comprehensive evaluations on two distinct medical datasets demonstrate state-of-the-art performance: TransUNet improves mDice/mIoU by 2.4%/4.2% on KiTS19, while SANet achieves 2.6%/3.5% gains on Polyps, highlighting its effectiveness and superiority. Code is available at GitHub.","short_abstract":"Medical image annotation is constrained by privacy concerns and labor-intensive labeling, significantly limiting the performance and generalization of segmentation models. While mask-controllable diffusion models excel in synthesis, they struggle with precise lesion-mask alignment. We propose \\textbf{Adaptively Distill...","url_abs":"https://arxiv.org/abs/2507.23652","url_pdf":"https://arxiv.org/pdf/2507.23652v1","authors":"[\"Kunpeng Qiu\",\"Zhiying Zhou\",\"Yongxin Guo\"]","published":"2025-07-31T15:32:06Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false}