{"ID":2880304,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.14779","arxiv_id":"2508.14779","title":"Hospital-Specific Bias in Patch-Based Pathology Models","abstract":"Pathology foundation models (PFMs) achieve strong performance on diverse histopathology tasks, but their sensitivity to hospital-specific domain shifts remains underexplored. We systematically evaluate state-of-the-art PFMs on TCGA patch-level datasets and introduce a lightweight adversarial adaptor to remove hospital-related domain information from latent representations. Experiments show that, while disease classification accuracy is largely maintained, the adaptor effectively reduces hospital-specific bias, as confirmed by t-SNE visualizations. Our study establishes a benchmark for assessing cross-hospital robustness in PFMs and provides a practical strategy for enhancing generalization under heterogeneous clinical settings. Our code is available at https://github.com/MengRes/pfm_domain_bias.","short_abstract":"Pathology foundation models (PFMs) achieve strong performance on diverse histopathology tasks, but their sensitivity to hospital-specific domain shifts remains underexplored. We systematically evaluate state-of-the-art PFMs on TCGA patch-level datasets and introduce a lightweight adversarial adaptor to remove hospital-...","url_abs":"https://arxiv.org/abs/2508.14779","url_pdf":"https://arxiv.org/pdf/2508.14779v2","authors":"[\"Mengliang Zhang\"]","published":"2025-08-20T15:25:16Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"eess.IV\"]","methods":"[]","has_code":false,"code_links":[{"ID":610667,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2880304,"paper_url":"https://arxiv.org/abs/2508.14779","paper_title":"Hospital-Specific Bias in Patch-Based Pathology Models","repo_url":"https://github.com/MengRes/pfm_domain_bias","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}