{"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/how-many-unicorns-are-in-this-image-a-safety","title":"How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs","arxiv_id":"2311.16101","date":"2023-11-27","proceeding":null,"authors":["Haoqin Tu","Chenhang Cui","Zijun Wang","Yiyang Zhou","Bingchen Zhao","Junlin Han","Wangchunshu Zhou","Huaxiu Yao","Cihang Xie"],"abstract":"This work focuses on the potential of Vision LLMs (VLLMs) in visual reasoning. Different from prior studies, we shift our focus from evaluating standard performance to introducing a comprehensive safety evaluation suite, covering both out-of-distribution (OOD) generalization and adversarial robustness. For the OOD evaluation, we present two novel VQA datasets, each with one variant, designed to test model performance under challenging conditions. In exploring adversarial robustness, we propose a straightforward attack strategy for misleading VLLMs to produce visual-unrelated responses. Moreover, we assess the efficacy of two jailbreaking strategies, targeting either the vision or language component of VLLMs. Our evaluation of 21 diverse models, ranging from open-source VLLMs to GPT-4V, yields interesting observations: 1) Current VLLMs struggle with OOD texts but not images, unless the visual information is limited; and 2) These VLLMs can be easily misled by deceiving vision encoders only, and their vision-language training often compromise safety protocols. We release this safety evaluation suite at https://github.com/UCSC-VLAA/vllm-safety-benchmark.","url_abs":"https://arxiv.org/abs/2311.16101v1","url_pdf":"https://arxiv.org/pdf/2311.16101v1.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":"how-many-unicorns-are-in-this-image-a-safety","repo_url":"https://github.com/ucsc-vlaa/vllm-safety-benchmark","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"adversarial-robustness","task_name":"Adversarial Robustness"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"},{"task_slug":"visual-reasoning","task_name":"Visual Reasoning"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2311.16101","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.16101"}},"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/ucsc-vlaa/vllm-safety-benchmark","reach":null}],"summary":{"ran_fixture":1,"ran":1},"by_repo_kind":{"official":{"samples":2,"ran":2,"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":2,"samples":[{"code_sha256_prefix":"ae36d350b447e2db","entry":"get_random_indices","repo":"ucsc-vlaa/vllm-safety-benchmark","repo_kind":"official","path":"safety_evaluations/ood_scenarios/evaluation.py","file_url":"https://github.com/ucsc-vlaa/vllm-safety-benchmark/blob/HEAD/safety_evaluations/ood_scenarios/evaluation.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ae36d350b447e2db"}},{"code_sha256_prefix":"9047aec079594943","entry":"prepare_eval_samples_and_dataset","repo":"ucsc-vlaa/vllm-safety-benchmark","repo_kind":"official","path":"safety_evaluations/ood_scenarios/evaluation.py","file_url":"https://github.com/ucsc-vlaa/vllm-safety-benchmark/blob/HEAD/safety_evaluations/ood_scenarios/evaluation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9047aec079594943"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}