{"ID":2832438,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2601.04199","arxiv_id":"2601.04199","title":"The Forgotten Shield: Safety Grafting in Parameter-Space for Medical MLLMs","abstract":"Medical Multimodal Large Language Models (Medical MLLMs) have achieved remarkable progress in specialized medical tasks; however, research into their safety has lagged, posing potential risks for real-world deployment. In this paper, we first establish a multidimensional evaluation framework to systematically benchmark the safety of current SOTA Medical MLLMs. Our empirical analysis reveals pervasive vulnerabilities across both general and medical-specific safety dimensions in existing models, particularly highlighting their fragility against cross-modality jailbreak attacks. Furthermore, we find that the medical fine-tuning process frequently induces catastrophic forgetting of the model's original safety alignment. To address this challenge, we propose a novel \"Parameter-Space Intervention\" approach for efficient safety re-alignment. This method extracts intrinsic safety knowledge representations from original base models and concurrently injects them into the target model during the construction of medical capabilities. Additionally, we design a fine-grained parameter search algorithm to achieve an optimal trade-off between safety and medical performance. Experimental results demonstrate that our approach significantly bolsters the safety guardrails of Medical MLLMs without relying on additional domain-specific safety data, while minimizing degradation to core medical performance.","short_abstract":"Medical Multimodal Large Language Models (Medical MLLMs) have achieved remarkable progress in specialized medical tasks; however, research into their safety has lagged, posing potential risks for real-world deployment. In this paper, we first establish a multidimensional evaluation framework to systematically benchmark...","url_abs":"https://arxiv.org/abs/2601.04199","url_pdf":"https://arxiv.org/pdf/2601.04199v1","authors":"[\"Jiale Zhao\",\"Xing Mou\",\"Jinlin Wu\",\"Hongyuan Yu\",\"Mingrui Sun\",\"Yang Shi\",\"Xuanwu Yin\",\"Zhen Chen\",\"Zhen Lei\",\"Yaohua Wang\"]","published":"2025-12-05T06:52:06Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\",\"cs.CL\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}