{"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/on-evaluating-adversarial-robustness-of-large","title":"On Evaluating Adversarial Robustness of Large Vision-Language Models","arxiv_id":"2305.16934","date":"2023-05-26","proceeding":"NeurIPS 2023 11","authors":["Yunqing Zhao","Tianyu Pang","Chao Du","Xiao Yang","Chongxuan Li","Ngai-Man Cheung","Min Lin"],"abstract":"Large vision-language models (VLMs) such as GPT-4 have achieved unprecedented performance in response generation, especially with visual inputs, enabling more creative and adaptable interaction than large language models such as ChatGPT. Nonetheless, multimodal generation exacerbates safety concerns, since adversaries may successfully evade the entire system by subtly manipulating the most vulnerable modality (e.g., vision). To this end, we propose evaluating the robustness of open-source large VLMs in the most realistic and high-risk setting, where adversaries have only black-box system access and seek to deceive the model into returning the targeted responses. In particular, we first craft targeted adversarial examples against pretrained models such as CLIP and BLIP, and then transfer these adversarial examples to other VLMs such as MiniGPT-4, LLaVA, UniDiffuser, BLIP-2, and Img2Prompt. In addition, we observe that black-box queries on these VLMs can further improve the effectiveness of targeted evasion, resulting in a surprisingly high success rate for generating targeted responses. Our findings provide a quantitative understanding regarding the adversarial vulnerability of large VLMs and call for a more thorough examination of their potential security flaws before deployment in practice. Code is at https://github.com/yunqing-me/AttackVLM.","url_abs":"https://arxiv.org/abs/2305.16934v2","url_pdf":"https://arxiv.org/pdf/2305.16934v2.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":"on-evaluating-adversarial-robustness-of-large","repo_url":"https://github.com/yunqing-me/attackvlm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"adversarial-robustness","task_name":"Adversarial Robustness"},{"task_slug":"response-generation","task_name":"Response Generation"},{"task_slug":"multimodal-generation","task_name":"multimodal generation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"blip","method_name":"BLIP"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"clip","method_name":"CLIP"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.16934","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.16934"}},"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/yunqing-me/attackvlm","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_honours":1,"ran_draft_wrong":2,"unverified":1},"by_repo_kind":{"official":{"samples":4,"ran":3,"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":0,"samples":[{"code_sha256_prefix":"2ee6e353f1dd4cf3","entry":"stable_diffusion_beta_schedule","repo":"yunqing-me/attackvlm","repo_kind":"official","path":"unidiff_tool/_eval_adv_img_i2t.py","file_url":"https://github.com/yunqing-me/attackvlm/blob/HEAD/unidiff_tool/_eval_adv_img_i2t.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2ee6e353f1dd4cf3"}},{"code_sha256_prefix":"948955a5b23f1937","entry":"to_tensor","repo":"yunqing-me/attackvlm","repo_kind":"official","path":"LAVIS_tool/_train_adv_img_query.py","file_url":"https://github.com/yunqing-me/attackvlm/blob/HEAD/LAVIS_tool/_train_adv_img_query.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"948955a5b23f1937"}},{"code_sha256_prefix":"30642020bc46ce62","entry":"unpreprocess","repo":"yunqing-me/attackvlm","repo_kind":"official","path":"unidiff_tool/_eval_adv_img_i2t.py","file_url":"https://github.com/yunqing-me/attackvlm/blob/HEAD/unidiff_tool/_eval_adv_img_i2t.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"30642020bc46ce62"}},{"code_sha256_prefix":"71dfa6ec4fceb241","entry":"find_max","repo":"yunqing-me/attackvlm","repo_kind":"official","path":"unidiff_tool/_train_trans_and_query_fixed_budget.py","file_url":"https://github.com/yunqing-me/attackvlm/blob/HEAD/unidiff_tool/_train_trans_and_query_fixed_budget.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"71dfa6ec4fceb241"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}