mlsanityData to 14 Oct 2026

FMM-Attack: A Flow-based Multi-modal Adversarial Attack on Video-based LLMs

arXiv 2403.13507 ↗First posted 20 Mar 2024cs.CVcited by 14

Not on the map. Why?

Abstract (arXiv)

Despite the remarkable performance of video-based large language models (LLMs), their adversarial threat remains unexplored. To fill this gap, we propose the first adversarial attack tailored for video-based LLMs by crafting flow-based multi-modal adversarial perturbations on a small fraction of frames within a video, dubbed FMM-Attack. Extensive experiments show that our attack can effectively induce video-based LLMs to generate incorrect answers when videos are added with imperceptible adversarial perturbations. Intriguingly, our FMM-Attack can also induce garbling in the model output, prompting video-based LLMs to hallucinate. Overall, our observations inspire a further understanding of multi-modal robustness and safety-related feature alignment across different modalities, which is of great importance for various large multi-modal models. Our code is available at https://github.com/THU-Kingmin/FMM-Attack.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 29 of its 37 references that are in the map. First posted on arXiv 20 Mar 2024.

Builds on

29 of 37 references.

References · 29

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  5. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  6. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  7. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  8. 2014CIDEr: Consensus-based Image Description Evaluationcited by 2,046
  9. 2023Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understandingcited by 1,471
  10. 2023Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Modelscited by 1,062
  11. 2023VideoChat: Chat-Centric Video Understandingcited by 857
  12. 2021CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrievalcited by 651
17 more
  1. 2023VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Taskscited by 501
  2. 2023FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Promptscited by 398
  3. 2020Multi-modal Transformer for Video Retrievalcited by 395
  4. 2024Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teacherscited by 393
  5. 2023On Evaluating Adversarial Robustness of Large Vision-Language Modelscited by 284
  6. 2021Advancing High-Resolution Video-Language Representation with Large-Scale Video Transcriptionscited by 226
  7. 2024Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Modelscited by 167
  8. 2015Sequence to Sequence -- Video to Textcited by 158
  9. 2023Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMscited by 124
  10. 2022End-to-end Generative Pretraining for Multimodal Video Captioningcited by 116
  11. 2021Bidirectional Projection Network for Cross Dimension Scene Understandingcited by 95
  12. 2021P2T: Pyramid Pooling Transformer for Scene Understandingcited by 84
  13. 2023A Pilot Study of Query-Free Adversarial Attack against Stable Diffusioncited by 76
  14. 2019Spatio-Temporal Dynamics and Semantic Attribute Enriched Visual Encoding for Video Captioningcited by 60
  15. 2018Sparse Adversarial Perturbations for Videoscited by 57
  16. 2021Open-book Video Captioning with Retrieve-Copy-Generate Networkcited by 43
  17. 2024On the Multi-modal Vulnerability of Diffusion Modelscited by 15

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 4; 2025: 7; 2026: 3.

14 papers.

Sort:

By area:

  1. 2024Visual Adversarial Attack on Vision-Language Models for Autonomous Drivingcited by 36
  2. 2025Survey of Adversarial Robustness in Multimodal Large Language Modelscited by 16
  3. 2024Video Watermarking: Safeguarding Your Video from (Unauthorized) Annotations by Video-based LLMscited by 7
  4. 2025Transferability of Adversarial Attacks in Video-based MLLMs: A Cross-modal Image-to-Video Approachcited by 7
  5. 2025Poisoning Prompt-Guided Sampling in Video Large Language Modelscited by 4
  6. 2026Adversarial Attacks on Multimodal Large Language Models: A Comprehensive Surveycited by 4
  7. 2025Investigating and Enhancing the Robustness of Large Multimodal Models Against Temporal Inconsistencycited by 2
  8. 2025Watch, Listen, Understand, Mislead: Tri-modal Adversarial Attacks on Short Videos for Content Appropriateness Evaluationcited by 2
  9. 2024Boundary-aware Decoupled Flow Networks for Realistic Extreme Rescalingcited by 1
  10. 2024Invertible Residual Rescaling Modelscited by 1
  11. 2025Investigating Vulnerabilities and Defenses Against Audio-Visual Attacks: A Comprehensive Survey Emphasizing Multimodal Modelscited by 1
  12. 2025FeatureFool: Zero-Query Fooling of Video Models via Feature Mapcited by 0
2 more
  1. 2026Jailbreaking Multimodal Large Language Models using Multi-Clip Videocited by 0
  2. 2026It Takes Little to Rewrite Perception: Targeted Semantic Substitution in Vision-Language Models at ε≤ 4/255cited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Do Input-Level Defenses Transfer to Observation-Level Attacks on VideoLLMs?
  2. 2024Feedback-based Modal Mutual Search for Attacking Vision-Language Pre-training Models
  3. 2025Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning
  4. 2023Misusing Tools in Large Language Models With Visual Adversarial Examples
  5. 2024Coercing LLMs to do and reveal (almost) anything
  6. 2025AdversariaLLM: A Unified and Modular Toolbox for LLM Robustness Research
  7. 2025From Pretrain to Pain: Adversarial Vulnerability of Video Foundation Models Without Task Knowledge
  8. 2024Breaking Down the Defenses: A Comparative Survey of Attacks on Large Language Models