mlsanityData to 7 Oct 2026

B-AVIBench: Towards Evaluating the Robustness of Large Vision-Language Model on Black-box Adversarial Visual-Instructions

arXiv 2403.09346 ↗First posted 14 Mar 2024cs.CV cs.AIcited by 26

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

Abstract (arXiv)

Large Vision-Language Models (LVLMs) have shown significant progress in responding well to visual-instructions from users. However, these instructions, encompassing images and text, are susceptible to both intentional and inadvertent attacks. Despite the critical importance of LVLMs' robustness against such threats, current research in this area remains limited. To bridge this gap, we introduce B-AVIBench, a framework designed to analyze the robustness of LVLMs when facing various Black-box Adversarial Visual-Instructions (B-AVIs), including four types of image-based B-AVIs, ten types of text-based B-AVIs, and nine types of content bias B-AVIs (such as gender, violence, cultural, and racial biases, among others). We generate 316K B-AVIs encompassing five categories of multimodal capabilities (ten tasks) and content bias. We then conduct a comprehensive evaluation involving 14 open-source LVLMs to assess their performance. B-AVIBench also serves as a convenient tool for practitioners to evaluate the robustness of LVLMs against B-AVIs. Our findings and extensive experimental results shed light on the vulnerabilities of LVLMs, and highlight that inherent biases exist even in advanced closed-source LVLMs like GeminiProVision and GPT-4V. This underscores the importance of enhancing the robustness, security, and fairness of LVLMs. The source code and benchmark are available at https://github.com/zhanghao5201/B-AVIBench.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 58 of its 75 references that are in the map. First posted on arXiv 14 Mar 2024.

Builds on

58 of 75 references.

Closest ancestors · 7

Same direction, newest first

  1. 2024One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models4 Mar
  2. 2024Robust CLIP: Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models19 Feb
  3. 2024Test-Time Backdoor Attacks on Multimodal Large Language Models13 Feb
  4. 2023On the Robustness of Large Multimodal Models Against Image Adversarial Attacks6 Dec
  5. 2023How Robust is Google's Bard to Adversarial Image Attacks?21 Sep
  6. 2023Visual Adversarial Examples Jailbreak Aligned Large Language Models22 Jun
  7. 2023On Evaluating Adversarial Robustness of Large Vision-Language Models26 May

Other references · 51

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2023Visual Instruction Tuningcited by 7,792
  6. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  7. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  8. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 2,914
  9. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,855
  10. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  11. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,105
  12. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,624
38 more
  1. 2015Microsoft COCO Captions: Data Collection and Evaluation Servercited by 1,447
  2. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
  3. 2023mPLUG-Owl: Modularization Empowers Large Language Models with Multimodalitycited by 1,038
  4. 2023ShareGPT4V: Improving Large Multi-Modal Models with Better Captionscited by 1,000
  5. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 979
  6. 2022A-OKVQA: A Benchmark for Visual Question Answering using World Knowledgecited by 702
  7. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 663
  8. 2023LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Modelcited by 602
  9. 2023Otter: A Multi-Modal Model with In-Context Instruction Tuningcited by 548
  10. 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 488
  11. 2017Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Modelscited by 459
  12. 2016Visual Dialogcited by 377
  13. 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 376
  14. 2018Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifierscited by 341
  15. 2023PandaGPT: One Model To Instruction-Follow Them Allcited by 330
  16. 2023Are aligned neural networks adversarially aligned?cited by 305
  17. 2023GPT4RoI: Instruction Tuning Large Language Model on Region-of-Interestcited by 300
  18. 2024MoE-LLaVA: Mixture of Experts for Large Vision-Language Modelscited by 298
  19. 2023InternLM-XComposer: A Vision-Language Large Model for Advanced Text-image Comprehension and Compositioncited by 297
  20. 2023LVLM-eHub: A Comprehensive Evaluation Benchmark for Large Vision-Language Modelscited by 217
  21. 2018Stress Test Evaluation for Natural Language Inferencecited by 211
  22. 2023LAMM: Language-Assisted Multi-Modal Instruction-Tuning Dataset, Framework, and Benchmarkcited by 169
  23. 2023Image Hijacks: Adversarial Images can Control Generative Models at Runtimecited by 165
  24. 2019Combating Adversarial Misspellings with Robust Word Recognitioncited by 162
  25. 2023Debiasing Vision-Language Models via Biased Promptscited by 147
  26. 2023On the Adversarial Robustness of Multi-Modal Foundation Modelscited by 142
  27. 2018Pathologies of Neural Models Make Interpretations Difficultcited by 140
  28. 2023How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMscited by 103
  29. 2018Benchmarking Neural Network Robustness to Common Corruptions and Surface Variationscited by 99
  30. 2023Breaking Common Sense: WHOOPS! A Vision-and-Language Benchmark of Synthetic and Compositional Imagescited by 79
  31. 2023VPGTrans: Transfer Visual Prompt Generator across LLMscited by 79
  32. 2018OpenTag: Open Attribute Value Extraction from Product Profiles [Deep Learning, Active Learning, Named Entity Recognition]cited by 62
  33. 2020SurFree: a fast surrogate-free black-box attackcited by 59
  34. 2023Vision-Language Models Performing Zero-Shot Tasks Exhibit Gender-based Disparitiescited by 31
  35. 2023Benchmarking Robustness of Adaptation Methods on Pre-trained Vision-Language Modelscited by 30
  36. 2024Benchmarking Large Multimodal Models against Common Corruptionscited by 22
  37. 2021Decision-based Black-box Attack Against Vision Transformers via Patch-wise Adversarial Removalcited by 16
  38. 2023Open-Vocabulary Animal Keypoint Detection with Semantic-feature Matchingcited by 10

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 10; 2025: 12; 2026: 4.

26 papers.

Sort:

By area:

  1. 2024A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trendscited by 82
  2. 2024T2VSafetyBench: Evaluating the Safety of Text-to-Video Generative Modelscited by 66
  3. 2025A Survey of Safety on Large Vision-Language Models: Attacks, Defenses and Evaluationscited by 66
  4. 2025Safety at Scale: A Comprehensive Survey of Large Model and Agent Safetycited by 63
  5. 2024Needle In A Multimodal Haystackcited by 56
  6. 2025Towards Robust and Secure Embodied AI: A Survey on Vulnerabilities and Attackscited by 32
  7. 2024VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Modelcited by 26
  8. 2025A Survey of Multimodal Hallucination Evaluation and Detectioncited by 26
  9. 2024Exploring Response Uncertainty in MLLMs: An Empirical Evaluation under Misleading Scenarioscited by 19
  10. 2025Survey of Adversarial Robustness in Multimodal Large Language Modelscited by 14
  11. 2025From LLMs to MLLMs to Agents: A Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystemcited by 11
  12. 2024Visual Riddles: a Commonsense and World Knowledge Challenge for Large Vision and Language Modelscited by 10
14 more
  1. 2025BBQ-V: Benchmarking Visual Stereotype Bias in Large Multimodal Modelscited by 10
  2. 2024Measuring the Measurers: Quality Evaluation of Hallucination Benchmarks for Large Vision-Language Modelscited by 8
  3. 2025Unveiling the Lack of LVLM Robustness to Fundamental Visual Variations: Why and Path Forwardcited by 7
  4. 2024Potential of Multimodal Large Language Models for Data Mining of Medical Images and Free-text Reportscited by 6
  5. 2025Losing the Plot: How VLM responses degrade on imperfect chartscited by 5
  6. 2025CBVLM: Training-free Explainable Concept-based Large Vision Language Models for Medical Image Classificationcited by 4
  7. 2025BLEnD-Vis: Benchmarking Multimodal Cultural Understanding in Vision Language Modelscited by 3
  8. 2026Making MLLMs Blind: Adversarial Smuggling Attacks in MLLM Content Moderationcited by 3
  9. 2024NEMO: Can Multimodal LLMs Identify Attribute-Modified Objects?cited by 2
  10. 2025On the Reliability of Vision-Language Models Under Adversarial Frequency-Domain Perturbationscited by 2
  11. 2024Efficient High-Resolution Visual Representation Learning with State Space Model for Human Pose Estimationcited by 0
  12. 2026When Relations Break: Analyzing Relation Hallucination in Vision-Language Model Under Rotation and Noisecited by 0
  13. 2026PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Modelscited by 0
  14. 2026Evolving Safety Landscape of Multi-modal Large Language Models: A Survey of Emerging Threats and Safeguardscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Towards Adversarially Robust Vision-Language Models: Insights from Design Choices and Prompt Formatting Techniques
  2. 2025Transferable Adversarial Attacks on Black-Box Vision-Language Models
  3. 2024Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey
  4. 2025Robust-LLaVA: On the Effectiveness of Large-Scale Robust Image Encoders for Multi-modal Large Language Models
  5. 2025Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness
  6. 2024Break the Visual Perception: Adversarial Attacks Targeting Encoded Visual Tokens of Large Vision-Language Models
  7. 2025VEAttack: Downstream-agnostic Vision Encoder Attack against Large Vision Language Models
  8. 2025Seeing the Threat: Vulnerabilities in Vision-Language Models to Adversarial Attack