{
  "id": 6850,
  "url": "https://arxiv.org/abs/2603.21904v1",
  "title": "SHAPE: Structure-aware Hierarchical Unsupervised Domain Adaptation with Plausibility Evaluation for Medical Image Segmentation",
  "summary": "Unsupervised Domain Adaptation (UDA) is essential for deploying medical segmentation models across diverse clinical environments. Existing methods are fundamentally limited, suffering from semantically unaware feature alignment that results in poor distributional fidelity and from pseudo-label validation that disregards global anatomical constraints, thus failing to prevent the formation of globally implausible structures. To address these issues, we propose SHAPE (Structure-aware Hierarchical U",
  "authors": "Linkuan Zhou, Yinghao Xia, Yufei Shen, Xiangyu Li, Wenjie Du, Cong Cong et al.",
  "category": "research",
  "topics": "safety-alignment,healthcare,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-03-23T12:25:56.000Z",
  "fetched_at": "2026-07-14T16:32:45.896Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/6850",
  "original_url": "https://arxiv.org/abs/2603.21904v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}