Lune

ACL Findings2025

VSCBench: Bridging the Gap in Vision-Language Model Safety Calibration

Jiahui Geng, Qing Li, Zongxiong Chen, Yuxia Wang, Derui Zhu, Zhuohan Xie, Chenyang Lyu, Xiuying Chen, Preslav Nakov, Fakhri Karray

Abstract

The rapid advancement of vision-language models (VLMs) has brought a lot of attention to their safety alignment. However, existing methods have primarily focused on model undersafety, where the model responds to hazardous queries, while neglecting oversafety, where the model refuses to answer safe queries. In this paper, we introduce the concept of safety calibration, which systematically addresses both undersafety and oversafety. Specifically, we present VSCBench, a novel dataset of 3,600 image-text pairs that are visually or textually similar but differ in terms of safety, which is designed to evaluate safety calibration across image-centric and text-centric scenarios. Based on our benchmark, we evaluate safety calibration across eleven widely used VLMs. Our extensive experiments revealed major issues with both undersafety and oversafety. We further investigated four approaches to improve the model's safety calibration. We found that even though some methods effectively calibrated the models' safety problems, these methods also lead to the degradation of models' utility. This trade-off underscores the urgent need for advanced calibration methods, and our benchmark provides a valuable tool for evaluating future approaches. Our code and data are available at https://github.com/ jiahuigeng/VSCBench.git.

Warning: This paper contains examples that are offensive or harmful in nature.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

1 top-tier paper has cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Direct Preference Optimization: Your Language Model is Secretly a Reward Model cites Language Models are Few-Shot Learners
  • Visual Instruction Tuning cites Language Models are Few-Shot Learners
  • Large Language Models are Zero-Shot Reasoners cites Language Models are Few-Shot Learners
  • Large Language Models are Zero-Shot Reasoners cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Con Instruction: Universal Jailbreaking of Multimodal Large Language Models via Non-Textual Modalities cites Visual Instruction Tuning
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers1

  1. Con Instruction: Universal Jailbreaking of Multimodal Large Language Models via Non-Textual ModalitiesJiahui Geng, Thy Thy Tran, Preslav Nakov, Iryna GurevychACL 2025

Builds on16

  1. Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
  2. Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
  3. Visual Instruction TuningHaotian Liu, Chunyuan Li, Qingyang Wu, Yong Jae LeeNeurIPS 2023 · 11,349 citations
  4. Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning et al.NeurIPS 2023 · 10,924 citations
  5. Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 citations

Related papers

  1. VLSBench: Unveiling Visual Leakage in Multimodal SafetyXuhao Hu, Dongrui Liu, Hao Li, Xuanjing Huang et al.ACL 2025
  2. HoliSafe: Holistic Safety Benchmarking and Modeling for Vision-Language ModelYoungwan Lee, Kangsan Kim, Kwanyong Park, Ilchae Jung, Soojin Jang, Seanie Lee, Yong-Ju Lee, Sung Ju HwangCVPR Findings 2026
  3. Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark StudyDongGeon Lee, Joonwon Jang, Jihae Jeong, Hwanjo YuEMNLP 2025 · 1 citation
  4. PII-VisBench: Evaluating Personally Identifiable Information Safety in Vision Language Models Along a Continuum of VisibilityG. M. Shahariar, Zabir Al Nazi, Md. Olid Hasan Bhuiyan, Zhouxing ShiACL Findings 2026
  5. When Helpers Become Hazards: A Benchmark for Analyzing Multimodal LLM-Powered Safety in Daily LifeXinyue Lou, Jinan Xu, Jingyi Yin, Xiaolong Wang et al.ACL Findings 2026 · 1 citation
Dusk over the sea between two cliffs drawn in fine vertical lines