mlsanityData to 14 Oct 2026

RapGuard: Safeguarding Multimodal Large Language Models via Rationale-aware Defensive Prompting

arXiv 2412.18826 ↗First posted 25 Dec 2024cs.CLcited by 19

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

also in 1 more
Abstract (arXiv)

While Multimodal Large Language Models (MLLMs) have made remarkable progress in vision-language reasoning, they are also more susceptible to producing harmful content compared to models that focus solely on text. Existing defensive prompting techniques rely on a static, unified safety guideline that fails to account for the specific risks inherent in different multimodal contexts. To address these limitations, we propose RapGuard, a novel framework that uses multimodal chain-of-thought reasoning to dynamically generate scenario-specific safety prompts. RapGuard enhances safety by adapting its prompts to the unique risks of each input, effectively mitigating harmful outputs while maintaining high performance on benign tasks. Our experimental results across multiple MLLM benchmarks demonstrate that RapGuard achieves state-of-the-art safety performance, significantly reducing harmful content without degrading the quality of responses.

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 40 of its 54 references that are in the map. First posted on arXiv 25 Dec 2024.

Builds on

40 of 54 references.

Closest ancestors · 8

Same direction, newest first

  1. 2024Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast13 Feb
  2. 2024Safety of Multimodal Large Language Models on Images and Texts1 Feb
  3. 2024MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance5 Jan
  4. 2023MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models29 Nov
  5. 2023FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts9 Nov
  6. 2023How Robust is Google's Bard to Adversarial Image Attacks?21 Sep
  7. 2023Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal Language Models26 Jul
  8. 2023Visual Adversarial Examples Jailbreak Aligned Large Language Models22 Jun

Other references · 32

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  3. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  4. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  5. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  6. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
  7. 2024Mixtral of Expertscited by 1,823
  8. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,652
  9. 2023Video-LLaVA: Learning United Visual Representation by Alignment Before Projectioncited by 1,402
  10. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,239
  11. 2023ShareGPT4V: Improving Large Multi-Modal Models with Better Captionscited by 1,005
  12. 2023A Survey on Multimodal Large Language Modelscited by 909
20 more
  1. 2023LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attentioncited by 781
  2. 2023CogVLM: Visual Expert for Pretrained Language Modelscited by 708
  3. 2023MiniGPT-v2: large language model as a unified interface for vision-language multi-task learningcited by 625
  4. 2023mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaborationcited by 547
  5. 2023LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignmentcited by 370
  6. 2023Set-of-Mark Prompting Unleashes Extraordinary Visual Grounding in GPT-4Vcited by 299
  7. 2024MM-LLMs: Recent Advances in MultiModal Large Language Modelscited by 278
  8. 2023Woodpecker: Hallucination Correction for Multimodal Large Language Modelscited by 226
  9. 2023Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attackscited by 194
  10. 2024Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Modelscited by 167
  11. 2022Multi-modal Molecule Structure-text Model for Text-based Retrieval and Editingcited by 166
  12. 2023Chain of Hindsight Aligns Language Models with Feedbackcited by 148
  13. 2023Mixture of Cluster-conditional LoRA Experts for Vision-language Instruction Tuningcited by 106
  14. 2023DRESS: Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedbackcited by 103
  15. 2023Safety Assessment of Chinese Large Language Modelscited by 85
  16. 2023A Challenger to GPT-4V? Early Explorations of Gemini in Visual Expertisecited by 49
  17. 2023Gaining Wisdom from Setbacks: Aligning Large Language Models via Mistake Analysiscited by 35
  18. 2023Aligning Large Language Models with Human: A Surveycited by 32
  19. 2024Mixture of insighTful Experts (MoTE): The Synergy of Thought Chains and Expert Mixtures in Self-Alignmentcited by 10
  20. 2023GPT-4V(ision) as A Social Media Analysis Enginecited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 15; 2026: 4.

19 papers.

Sort:

By area:

  1. 2025Towards Trustworthy GUI Agents: A Surveycited by 34
  2. 2025HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden Statescited by 33
  3. 2025MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMscited by 33
  4. 2025Can't See the Forest for the Trees: Benchmarking Multimodal Safety Awareness for Multimodal LLMscited by 23
  5. 2025Equilibrate RLHF: Towards Balancing Helpfulness-Safety Trade-off in Large Language Modelscited by 22
  6. 2025Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Studycited by 12
  7. 2025Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025cited by 12
  8. 2025USB: A Comprehensive and Unified Safety Evaluation Benchmark for Multimodal Large Language Modelscited by 10
  9. 2025SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Modelscited by 10
  10. 2025DAVSP: Safety Alignment for Large Vision-Language Models via Deep Aligned Visual Safety Promptcited by 8
  11. 2025Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Modelscited by 7
  12. 2025MSR-Align: Policy-Grounded Multimodal Alignment for Safety-Aware Reasoning in Vision-Language Modelscited by 7
7 more
  1. 2025QuadSentinel: Sequent Safety for Machine-Checkable Control in Multi-agent Systemscited by 4
  2. 2026Physical Prompt Injection Attacks on Large Vision-Language Modelscited by 4
  3. 2026Visual Self-Fulfilling Alignment: Shaping Safety-Oriented Personas via Threat-Related Imagescited by 3
  4. 2025A Survey on Training-free Alignment of Large Language Modelscited by 2
  5. 2026Safe responses matter: Output-aware safety guardrail mitigate over-refusal in MLLMscited by 2
  6. 2025LLaVAShield: Safeguarding Multimodal Multi-Turn Dialogues in Vision-Language Modelscited by 1
  7. 2026Dictionary-Aligned Concept Control for Safeguarding Multimodal LLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026ThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models
  2. 2024Multimodal Situational Safety
  3. 2025VLMGuard-R1: Proactive Safety Alignment for VLMs via Reasoning-Driven Prompt Optimization
  4. 2025Think-Reflect-Revise: A Policy-Guided Reflective Framework for Safety Alignment in Large Vision Language Models
  5. 2025Automating Steering for Safe Multimodal Large Language Models
  6. 2026Evolving Contextual Safety in Multi-Modal Large Language Models via Inference-Time Self-Reflective Memory
  7. 2026MME-Safety: A Fine-grained Benchmark for Safety Evaluation of MLLMs
  8. 2025ProGuard: Towards Proactive Multimodal Safeguard