{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/vlsbench-unveiling-visual-leakage-in","title":"VLSBench: Unveiling Visual Leakage in Multimodal Safety","arxiv_id":"2411.19939","date":"2024-11-29","proceeding":null,"authors":["Xuhao Hu","Dongrui Liu","Hao Li","Xuanjing Huang","Jing Shao"],"abstract":"Safety concerns of Multimodal large language models (MLLMs) have gradually become an important problem in various applications. Surprisingly, previous works indicate a counterintuitive phenomenon that using textual unlearning to align MLLMs achieves comparable safety performances with MLLMs aligned with image text pairs. To explain such a phenomenon, we discover a Visual Safety Information Leakage (VSIL) problem in existing multimodal safety benchmarks, i.e., the potentially risky content in the image has been revealed in the textual query. Thus, MLLMs can easily refuse these sensitive image-text pairs according to textual queries only, leading to unreliable cross-modality safety evaluation of MLLMs. We also conduct a further comparison experiment between textual alignment and multimodal alignment to highlight this drawback. To this end, we construct multimodal Visual Leakless Safety Bench (VLSBench) with 2.2k image-text pairs through an automated data pipeline. Experimental results indicate that VLSBench poses a significant challenge to both open-source and close-source MLLMs, e.g., LLaVA, Qwen2-VL and GPT-4o. Besides, we empirically compare textual and multimodal alignment methods on VLSBench and find that textual alignment is effective enough for multimodal safety scenarios with VSIL, while multimodal alignment is preferable for safety scenarios without VSIL. Code and data are released under https://github.com/AI45Lab/VLSBench","url_abs":"https://arxiv.org/abs/2411.19939v3","url_pdf":"https://arxiv.org/pdf/2411.19939v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"vlsbench-unveiling-visual-leakage-in","repo_url":"https://github.com/ai45lab/vlsbench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"align","method_name":"ALIGN"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2411.19939","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2411.19939"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ai45lab/vlsbench","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":1,"unverified":5},"by_repo_kind":{"official":{"samples":6,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":6,"samples":[{"code_sha256_prefix":"f41cb1a19b154297","entry":"encode_image","repo":"ai45lab/vlsbench","repo_kind":"official","path":"eval_utils.py","file_url":"https://github.com/ai45lab/vlsbench/blob/HEAD/eval_utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f41cb1a19b154297"}},{"code_sha256_prefix":"f6c3e0ef12ef5779","entry":"find_files_with_name","repo":"ai45lab/vlsbench","repo_kind":"official","path":"eval_utils.py","file_url":"https://github.com/ai45lab/vlsbench/blob/HEAD/eval_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f6c3e0ef12ef5779"}},{"code_sha256_prefix":"8176c7a3579cba1d","entry":"guess_image_type_from_base64","repo":"ai45lab/vlsbench","repo_kind":"official","path":"eval_utils.py","file_url":"https://github.com/ai45lab/vlsbench/blob/HEAD/eval_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8176c7a3579cba1d"}},{"code_sha256_prefix":"4552e857b0cf7ef7","entry":"load_image","repo":"ai45lab/vlsbench","repo_kind":"official","path":"models/load_llava.py","file_url":"https://github.com/ai45lab/vlsbench/blob/HEAD/models/load_llava.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4552e857b0cf7ef7"}},{"code_sha256_prefix":"99d66c86efac69ff","entry":"load_images","repo":"ai45lab/vlsbench","repo_kind":"official","path":"models/load_llava.py","file_url":"https://github.com/ai45lab/vlsbench/blob/HEAD/models/load_llava.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"99d66c86efac69ff"}},{"code_sha256_prefix":"9aa7187541e084e2","entry":"load_images","repo":"ai45lab/vlsbench","repo_kind":"official","path":"models/utils.py","file_url":"https://github.com/ai45lab/vlsbench/blob/HEAD/models/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9aa7187541e084e2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}