Enhancing the Safety of Medical Vision-Language Models by Synthetic Demonstrations

[2506.09067] Enhancing the Safety of Medical Vision-Language Models by Synthetic Demonstrations Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Computer Vision and Pattern Recognition arXiv:2506.09067 (cs) [Submitted on 8 Jun 2025 ( v1 ), last revised 9 Apr 2026 (this version, v2)] Title: Enhancing the Safety of Medical Vision-Language Models by Synthetic Demonstrations Authors: Zhiyu Xue , Reza Abbasi-Asl , Ramtin Pedarsani View a PDF of the paper titled Enhancing the Safety of Medical Vision-Language Models by Synthetic Demonstrations, by Zhiyu Xue and 2 other authors View PDF HTML (experimental) Abstract: Generative medical vision-language models~(Med-VLMs) are primarily designed to generate complex textual information~(e.g., diagnostic reports) from multimodal inputs including vision modality~(e.g., medical images) and language modality~(e.g., clinical queries). However, their security vulnerabilities remain underexplored. Med-VLMs should be capable of rejecting harmful queries, such as \textit{Provide detailed instructions for using this CT scan for insurance fraud}. At the same time, addressing security concerns introduces the risk of over-defense, where safety-enhancing mechanisms may degrade general performance, causing Med-VLMs to reject benign clinical queries. In this paper, we propose a novel inference-time defense strategy to mitigate harmful queries, enabling defense against visual and textual jailbreak attacks. Using diverse medical imaging datasets collected from nine modalities, we demonstrate that our defense strategy based on synthetic clinical demonstrations enhances model safety without significantly compromising performance. Additionally, we find that increasing the demonstration budget alleviates the over-defense issue. We then introduce a mixed demonstration strategy as a trade-off solution for balancing security and performance under few-shot demonstration budget constraints. Subjects: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2506.09067 [cs.CV] (or arXiv:2506.09067v2 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2506.09067 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Zhiyu Xue [ view email ] [v1] Sun, 8 Jun 2025 16:26:51 UTC (1,965 KB) [v2] Thu, 9 Apr 2026 22:50:08 UTC (2,000 KB) Full-text links: Access Paper: View a PDF of the paper titled Enhancing the Safety of Medical Vision-Language Models by Synthetic Demonstrations, by Zhiyu Xue and 2 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CV < prev | next > new | recent | 2025-06 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• Are Vision-Language-Action Models Robust to One-Step Observation Perturbations?#1116881
• Uncertainty Quantification for Flow-Based Vision-Language-Action Models#282795
• Prompt injection attacks on vision-language models for surgical decision support#406906
• Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms#141781
• Evidence-Gated Task and Motion Planning with Vision-Language Models#471287