Cultural Safety Evaluation of VLMs

VLM: Vision-Language Model

Follow this topic Ask about this topic

Latest papers

14 papers3 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption Bank
Oct 7, 2026Samuel Tetteh, Cody FlemingCultural Safety Evaluation of VLMs VLM Evaluation
Iowa State University Ames, Iowa, USA
From Local Evidence to Safety Verdicts: Causal Tracing in Vision-Language Models
Oct 5, 2026Faezeh Dehghan Tarzjani, Mevan Wijewardena, Alexander Romanus +1Cultural Safety Evaluation of VLMs VLM Safety
Department of Electrical and Computer Engineering University of Southern California, Los Angeles, CA, USA · Department of Computer Science University of Southern California, Los Angeles, CA, USA
Towards Systematic Qualification of Vision-Language Models for Automotive Perception Systems
Sep 22, 2026Malsha Ashani Mahawatta Dona, Konstantinos Rokanas, Alexander Säfström +2Cultural Safety Evaluation of VLMs VLM Evaluation
University of Gothenburg and Chalmers University of Technology, Gothenburg, Sweden
Knowing What Not to Answer: Selective Non-Compliance in Vision-Language Models
Sep 4, 2026Minji Kim, Jihyoung Jang, Hyounghun KimCultural Safety Evaluation of VLMs VLM Evaluation
Graduate School of Artificial Intelligence, POSTECH · Department of Computer Science and Engineering, POSTECH
Beyond the Verdict: Evidence-Aligned Evaluation of Visual Prompt-Injection Guardrails
Sep 2, 2026Suyoung Lee, Myungsub ChoiCultural Safety Evaluation of VLMs VLM Evaluation
Tynapse, Seoul, Republic of Korea
EviSafe: Evidence-Grounded Safety Evaluation for Vision-Language Models
Aug 24, 2026Xuetong Li, Gaofeng LiuCultural Safety Evaluation of VLMs VLM Safety
School of Mathematical Sciences, MOE Key Lab of Scientific and Engineering Computing, Shanghai Center for Applied Mathematics, Shanghai Jiao Tong University · Department of Automation, Shanghai Jiao Tong University
Pluralis v0.1: Towards a Multicultural, Multimodal, Multilingual Benchmark for AI Risk and Reliability
Jul 7, 2026Alicia Parrish, Rajat Shinde, Sanket Badhe +57VLM Evaluation Cultural Safety Evaluation of VLMs
Mechanistically Interpreting Compression in Vision-Language Models
Mar 26, 2026Veeraraju Elluru, Arth Singh, Roberto Aguero +3Cultural Safety Evaluation of VLMs LVLM Pruning
Indian Institute of Technology Jodhpur · AIM Intelligence · National Institute of Technology Agartala +3
TreeTeaming: Autonomous Red-Teaming of Vision-Language Models via Hierarchical Strategy Exploration
Mar 24, 2026Chunxiao Li, Lijun Li, Jing ShaoCultural Safety Evaluation of VLMs Automated Red Teaming
Shanghai Artificial Intelligence Laboratory
SAVeS: Steering Safety Judgments in Vision-Language Models via Semantic Cues
Mar 19, 2026Carlos Hinojosa, Clemens Grange, Bernard GhanemCultural Safety Evaluation of VLMs Vision-Language Models
King Abdullah University of Science and Technology (KAUST), Saudi Arabia · Technical University of Munich (TUM), Germany
HomeSafe-Bench: Evaluating Vision-Language Models on Unsafe Action Detection for Embodied Agents in Household Scenarios
Mar 12, 2026Jiayue Pu, Zhongxiang Sun, Zilu Zhang +2Cultural Safety Evaluation of VLMs AI Agent Safety
Gaoling School of Artificial Intelligence, Renmin University of China · University of Chinese Academy of Sciences · Beijing University of Posts and Telecommunications
SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety
Mar 3, 2026Zixuan Xu, Tiancheng He, Huahui Yi +7Cultural Safety Evaluation of VLMs VLM Safety
Huazhong University of Science and Technology · Beijing University of Posts and Telecommunications · West China Hospital, Sichuan University +1
Towards Policy-Adaptive Image Guardrail: Benchmark and Method
Mar 1, 2026Caiyong Piao, Zhiyuan Yan, Haoming Xu +4Cultural Safety Evaluation of VLMs VLM Evaluation
Fudan University · Tencent · Peking University
MiSCHiEF: A Benchmark in Minimal-Pairs of Safety and Culture for Holistic Evaluation of Fine-Grained Image-Caption Alignment
Feb 21, 2026Sagarika Banerjee, Tangatar Madi, Advait Swaminathan +4Cultural Safety Evaluation of VLMs VLM Evaluation
Algoverse AI Research · PocketFM

Oct 7, 2026J/K move · Enter open · S save

It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption Bank

Samuel Tetteh, Cody Fleming

Iowa State University Ames, Iowa, USA

Cultural Safety Evaluation of VLMs VLM Evaluation VLM Robustness Counterfactual Evaluation of VLMs Vision-Language Alignment

Open paper AskPDF arXiv

Frozen vision-language models increasingly provide safety signals for reinforcement learning. Their use assumes that similarity to language describing danger indicates the hazard itself. Yet policy return and collision rate cannot reveal whether a score detects hazards or responds to correlated features of the scene. VLM-based methods have reported gains in driving and safe-RL benchmarks by converting image-text similarity into rewards, costs, or confidence weights. Such signals promise to reduce reliance on manually designed feedback. They may also reflect prompt structure, embedding geometry, or camera viewpoint, leaving their safety meaning unverified. To address this gap, we present a controlled evaluation of a frozen CLIP prompt-margin safety score. We apply the score to trajectories generated by policies that never receive it, match pre-contact observations to contact-free observations with comparable hazard geometry, and vary the captions, encoder, and camera view. Across three policies, 180 episodes, and 130 isolated contact onsets, the score decreases for about twenty steps before contact. Mechanism controls indicate that the score mainly tracks resemblance to the scene shared by its captions and changes with caption separation and camera view. A constant-confidence control retains the lower catastrophe-rate point estimate, so policy gains do not establish hazard perception.