mlsanityData to 13 Oct 2026

OmniSafeBench-MM: A Unified Benchmark and Toolbox for Multimodal Jailbreak Attack-Defense Evaluation

arXiv 2512.06589 ↗First posted 6 Dec 2025cs.CR cs.CVcited by 10

On the map: LLM safety and security → Jailbreaking

Abstract (arXiv)

Recent advances in multi-modal large language models (MLLMs) have enabled unified perception-reasoning capabilities, yet these systems remain highly vulnerable to jailbreak attacks that bypass safety alignment and induce harmful behaviors. Existing benchmarks such as JailBreakV-28K, MM-SafetyBench, and HADES provide valuable insights into multi-modal vulnerabilities, but they typically focus on limited attack scenarios, lack standardized defense evaluation, and offer no unified, reproducible toolbox. To address these gaps, we introduce OmniSafeBench-MM, which is a comprehensive toolbox for multi-modal jailbreak attack-defense evaluation. OmniSafeBench-MM integrates 13 representative attack methods, 15 defense strategies, and a diverse dataset spanning 9 major risk domains and 50 fine-grained categories, structured across consultative, imperative, and declarative inquiry types to reflect realistic user intentions. Beyond data coverage, it establishes a three-dimensional evaluation protocol measuring (1) harmfulness, distinguished by a granular, multi-level scale ranging from low-impact individual harm to catastrophic societal threats, (2) intent alignment between responses and queries, and (3) response detail level, enabling nuanced safety-utility analysis. We conduct extensive experiments on 10 open-source and 8 closed-source MLLMs to reveal their vulnerability to multi-modal jailbreak. By unifying data, methodology, and evaluation into an open-source, reproducible platform, OmniSafeBench-MM provides a standardized foundation for future research. The code is released at https://github.com/jiaxiaojunQAQ/OmniSafeBench-MM.

Key facts. Cited by 10 papers in the map (data through 13 Oct 2026). Placed in Jailbreaking, one of 12 directions in LLM safety and security. Builds on 46 of its 54 references that are in the map. First posted on arXiv 6 Dec 2025.

Builds on

46 of 54 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy26 Mar
  2. 2024Bag of Tricks: Benchmarking of Jailbreak Attacks on LLMs13 Jun
  3. 2024Improved Techniques for Optimization-Based Jailbreaking on Large Language Models31 May
  4. 2024JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks3 Apr

Other references · 42

  1. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,379
  2. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,363
  3. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,703
  4. 2023FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Promptscited by 394
  5. 2023Visual Adversarial Examples Jailbreak Aligned Large Language Modelscited by 326
  6. 2023A Survey on Multimodal Large Language Models for Autonomous Drivingcited by 290
  7. 2023MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Modelscited by 272
  8. 2024Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Modelscited by 166
  9. 2024Jailbreaking Attack against Multimodal Large Language Modelcited by 134
  10. 2024Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Modelscited by 133
  11. 2024AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shield Promptingcited by 125
  12. 2024Eyes Closed, Safety On: Protecting Multimodal LLMs via Image-to-Text Transformationcited by 124
30 more
  1. 2024MLLM-Protector: Ensuring MLLM's Safety without Hurting Performancecited by 117
  2. 2024Jailbreak Vision Language Models via Bi-Modal Adversarial Promptcited by 109
  3. 2024Natural Language Understanding and Inference with MLLM in Visual Question Answering: A Surveycited by 76
  4. 2024Visual-RolePlay: Universal Jailbreak Attack on MultiModal Large Language Models via Role-playing Image Charactercited by 66
  5. 2024Jailbreak Large Vision-Language Models Through Multi-Modal Linkagecited by 61
  6. 2024White-box Multimodal Jailbreaks Against Large Vision-Language Modelscited by 59
  7. 2024LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Modelscited by 52
  8. 2024ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectorscited by 51
  9. 2025Jailbreaking Multimodal Large Language Models via Shuffle Inconsistencycited by 50
  10. 2025Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignmentcited by 43
  11. 2025GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoningcited by 42
  12. 2024Perception-guided Jailbreak against Text-to-Image Modelscited by 40
  13. 2025Distraction is All You Need for Multimodal Large Language Model Jailbreakingcited by 36
  14. 2025HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden Statescited by 33
  15. 2024Heuristic-Induced Multimodal Risk Distribution Jailbreak Attack for Multimodal Large Language Modelscited by 32
  16. 2024CoCA: Regaining Safety-awareness of Multimodal Large Language Models with Constitutional Calibrationcited by 31
  17. 2024Cross-modality Information Check for Detecting Jailbreaking in Multimodal Large Language Modelscited by 25
  18. 2025Visual Contextual Attack: Jailbreaking MLLMs with Image-Driven Context Injectioncited by 24
  19. 2024UniGuard: Towards Universal Safety Guardrails for Jailbreak Attacks on Multimodal Large Language Modelscited by 19
  20. 2025Oyster-I: Beyond Refusal -- Constructive Safety Alignment for Responsible Language Modelscited by 19
  21. 2024PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximizationcited by 17
  22. 2025VisCRA: A Visual Chain Reasoning Attack for Jailbreaking Multimodal Large Language Modelscited by 12
  23. 2025JPS: Jailbreak Multimodal Large Language Models with Collaborative Visual Perturbation and Textual Steeringcited by 10
  24. 2025Align is not Enough: Multimodal Universal Jailbreak Attack against Multimodal Large Language Modelscited by 9
  25. 2024Defending LVLMs Against Vision Attacks through Partial-Perception Supervisioncited by 7
  26. 2025`From Prompt to Perturbation': An Adaptive Framework for Voice-Based Jailbreaks on Audio LLMscited by 7
  27. 2025Implicit Jailbreak Attacks via Cross-Modal Information Concealment on Vision-Language Modelscited by 6
  28. 2025Cross-Modal Obfuscation for Jailbreak Attacks on Large Vision-Language Modelscited by 6
  29. 2025QGuard:Question-based Zero-shot Guard for Multi-modal LLM Safetycited by 4
  30. 2025PolyJailbreak: Cross-Modal Jailbreaking Attacks on Black-Box Multimodal LLMscited by 3

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 1; 2026: 9.

10 papers.

  1. 2026SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agentscited by 22
  2. 2026OpenRT: An Open-Source Red Teaming Framework for Multimodal LLMscited by 20
  3. 2026Obscure but Effective: Classical Chinese Jailbreak Prompt Optimization via Bio-Inspired Searchcited by 12
  4. 2026Knowledge-Driven Multi-Turn Jailbreaking on Large Language Modelscited by 7
  5. 2026Visual Self-Fulfilling Alignment: Shaping Safety-Oriented Personas via Threat-Related Imagescited by 2
  6. 2025Casting a SPELL: Sentence Pairing Exploration for LLM Limitation-breakingcited by 1
  7. 2026Adversarial Attacks Against MLLMs via Progressive Resolution Processing and Adaptive Feature Alignmentcited by 1
  8. 2026PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Modelscited by 0
  9. 2026MMJailBench: A Factorized Benchmark for Disentangling Multimodal Jailbreak Vulnerabilitiescited by 0
  10. 2026MME-Safety: A Fine-grained Benchmark for Safety Evaluation of MLLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts Against Large Language Models
  2. 2026JailMeter: An Evidence-Based Evaluation Framework for Jailbreak Attacks on Large Language Models
  3. 2025Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience
  4. 2024JailbreakLens: Visual Analysis of Jailbreak Attacks Against Large Language Models
  5. 2024JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models
  6. 2024AttackEval: How to Evaluate the Effectiveness of Jailbreak Attacking on Large Language Models
  7. 2026How Jailbreak Attacks Inform Safety Alignment: A Defender-Centric, Shapley-Based Evaluation of Jailbreak Contributions
  8. 2025Speak Easy: Eliciting Harmful Jailbreaks from LLMs with Simple Interactions