{"ID":2868024,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.16886","arxiv_id":"2509.16886","title":"SAM-DCE: Addressing Token Uniformity and Semantic Over-Smoothing in Medical Segmentation","abstract":"The Segment Anything Model (SAM) demonstrates impressive zero-shot segmentation ability on natural images but encounters difficulties in medical imaging due to domain shifts, anatomical variability, and its reliance on user-provided prompts. Recent prompt-free adaptations alleviate the need for expert intervention, yet still suffer from limited robustness and adaptability, often overlooking the issues of semantic over-smoothing and token uniformity. We propose SAM-DCE, which balances local discrimination and global semantics while mitigating token uniformity, enhancing inter-class separability, and enriching mask decoding with fine-grained, consistent representations. Extensive experiments on diverse medical benchmarks validate its effectiveness.","short_abstract":"The Segment Anything Model (SAM) demonstrates impressive zero-shot segmentation ability on natural images but encounters difficulties in medical imaging due to domain shifts, anatomical variability, and its reliance on user-provided prompts. Recent prompt-free adaptations alleviate the need for expert intervention, yet...","url_abs":"https://arxiv.org/abs/2509.16886","url_pdf":"https://arxiv.org/pdf/2509.16886v2","authors":"[\"Yingzhen Hu\",\"Yiheng Zhong\",\"Ruobing Li\",\"Yingxue Su\",\"Jiabao An\",\"Feilong Tang\",\"Jionglong Su\",\"Imran Razzak\"]","published":"2025-09-21T02:39:48Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}