{"ID":2852859,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.17687","arxiv_id":"2510.17687","title":"CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks","abstract":"Multimodal Large Language Models (MLLMs) achieve strong reasoning and perception capabilities but are increasingly vulnerable to jailbreak attacks. While existing work focuses on explicit attacks, where malicious content resides in a single modality, recent studies reveal implicit attacks, in which benign text and image inputs jointly express unsafe intent. Such joint-modal threats are difficult to detect and remain underexplored, largely due to the scarcity of high-quality implicit data. We propose ImpForge, an automated red-teaming pipeline that leverages reinforcement learning with tailored reward modules to generate diverse implicit samples across 14 domains. Building on this dataset, we further develop CrossGuard, an intent-aware safeguard providing robust and comprehensive defense against both explicit and implicit threats. Extensive experiments across safe and unsafe benchmarks, implicit and explicit attacks, and multiple out-of-domain settings demonstrate that CrossGuard significantly outperforms existing defenses, including advanced MLLMs and guardrails, achieving stronger security while maintaining high utility. This offers a balanced and practical solution for enhancing MLLM robustness against real-world multimodal threats. Our code is released: https://github.com/ZhangXu0963/CrossGuard.","short_abstract":"Multimodal Large Language Models (MLLMs) achieve strong reasoning and perception capabilities but are increasingly vulnerable to jailbreak attacks. While existing work focuses on explicit attacks, where malicious content resides in a single modality, recent studies reveal implicit attacks, in which benign text and imag...","url_abs":"https://arxiv.org/abs/2510.17687","url_pdf":"https://arxiv.org/pdf/2510.17687v2","authors":"[\"Xu Zhang\",\"Hao Li\",\"Zhichao Lu\"]","published":"2025-10-20T16:02:34Z","proceeding":"cs.CR","tasks":"[\"cs.CR\",\"cs.AI\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":608034,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2852859,"paper_url":"https://arxiv.org/abs/2510.17687","paper_title":"CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks","repo_url":"https://github.com/ZhangXu0963/CrossGuard","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}