mlsanityData to 14 Oct 2026

Adversarial Robustness for Visual Grounding of Multimodal Large Language Models

arXiv 2405.09981 ↗First posted 16 May 2024cs.CVcited by 23

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

Abstract (arXiv)

Multi-modal Large Language Models (MLLMs) have recently achieved enhanced performance across various vision-language tasks including visual grounding capabilities. However, the adversarial robustness of visual grounding remains unexplored in MLLMs. To fill this gap, we use referring expression comprehension (REC) as an example task in visual grounding and propose three adversarial attack paradigms as follows. Firstly, untargeted adversarial attacks induce MLLMs to generate incorrect bounding boxes for each object. Besides, exclusive targeted adversarial attacks cause all generated outputs to the same target bounding box. In addition, permuted targeted adversarial attacks aim to permute all bounding boxes among different objects within a single image. Extensive experiments demonstrate that the proposed methods can successfully attack visual grounding capabilities of MLLMs. Our methods not only provide a new perspective for designing novel attacks but also serve as a strong baseline for improving the adversarial robustness for visual grounding of MLLMs.

Key facts. Cited by 23 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 42 of its 46 references that are in the map. First posted on arXiv 16 May 2024.

Builds on

42 of 46 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images20 Jan
  2. 2023BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP26 Nov
  3. 2023How Robust is Google's Bard to Adversarial Image Attacks?21 Sep
  4. 2023Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs19 Jul
  5. 2023On Evaluating Adversarial Robustness of Large Vision-Language Models26 May

Other references · 37

  1. 2023Visual Instruction Tuningcited by 7,869
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  5. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  6. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,322
  7. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  8. 2021Align before Fuse: Vision and Language Representation Learning with Momentum Distillationcited by 1,619
  9. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  10. 2023Kosmos-2: Grounding Multimodal Large Language Models to the Worldcited by 1,089
  11. 2016Modeling Context in Referring Expressionscited by 1,063
  12. 2023Shikra: Unleashing Multimodal LLM's Referential Dialogue Magiccited by 906
25 more
  1. 2015Generation and Comprehension of Unambiguous Object Descriptionscited by 904
  2. 2023MiniGPT-v2: large language model as a unified interface for vision-language multi-task learningcited by 625
  3. 2018Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networkscited by 519
  4. 2018Black-box Adversarial Attacks with Limited Queries and Informationcited by 501
  5. 2023VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Taskscited by 501
  6. 2019On Evaluating Adversarial Robustnesscited by 447
  7. 2020Backdoor Learning: A Surveycited by 426
  8. 2023Are aligned neural networks adversarially aligned?cited by 315
  9. 2023Follow Your Pose: Pose-Guided Text-to-Video Generation using Pose-Free Videoscited by 314
  10. 2021VisualGPT: Data-efficient Adaptation of Pretrained Language Models for Image Captioningcited by 191
  11. 2021Referring Transformer: A One-step Approach to Multi-task Visual Groundingcited by 143
  12. 2019TBT: Targeted Neural Network Attack with Bit Trojancited by 113
  13. 2017Grounding Referring Expressions in Images by Variational Contextcited by 82
  14. 2021Improving Adversarial Robustness via Channel-wise Activation Suppressingcited by 76
  15. 2021Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bitscited by 49
  16. 2023Backdoor Defense via Adaptively Splitting Poisoned Datasetcited by 43
  17. 2020Targeted Attack for Deep Hashing based Retrievalcited by 35
  18. 2020Improving Query Efficiency of Black-box Adversarial Attackcited by 34
  19. 2024Stop Reasoning! When Multimodal LLM with Chain-of-Thought Reasoning Meets Adversarial Imagecited by 26
  20. 2022Imperceptible and Robust Backdoor Attack in 3D Point Cloudcited by 25
  21. 2022Hardly Perceptible Trojan Attack against Neural Networks with Bit Flipscited by 22
  22. 2022SimVTP: Simple Video Text Pre-training with Masked Autoencoderscited by 21
  23. 2024On the Multi-modal Vulnerability of Diffusion Modelscited by 15
  24. 2024Energy-Latency Manipulation of Multi-modal Large Language Models via Verbose Samplescited by 14
  25. 2022Versatile Weight Attack via Flipping Limited Bitscited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 7; 2025: 13; 2026: 3.

23 papers.

Sort:

By area:

  1. 2024A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trendscited by 84
  2. 2025A Survey of Safety on Large Vision-Language Models: Attacks, Defenses and Evaluationscited by 67
  3. 2024One Perturbation is Enough: On Generating Universal Adversarial Perturbations against Vision-Language Pre-training Modelscited by 40
  4. 2025Visual Large Language Models for Generalized and Specialized Applicationscited by 36
  5. 2025Towards Robust and Secure Embodied AI: A Survey on Vulnerabilities and Attackscited by 34
  6. 2025ADVEDM:Fine-grained Adversarial Attack against VLM-based Embodied Agentscited by 23
  7. 2024CLIP-Guided Generative Networks for Transferable Targeted Adversarial Attackscited by 19
  8. 2024Semantic-Aligned Adversarial Evolution Triangle for High-Transferability Vision-Language Attackcited by 18
  9. 2025Mirage in the Eyes: Hallucination Attack on Multi-modal Large Language Models with Only Attention Sinkcited by 15
  10. 2025On the Robustness of GUI Grounding Models Against Image Attackscited by 14
  11. 2025V-Attack: Targeting Disentangled Value Features for Controllable Adversarial Attacks on LVLMscited by 9
  12. 2024Video Watermarking: Safeguarding Your Video from (Unauthorized) Annotations by Video-based LLMscited by 7
11 more
  1. 2024Systematic Categorization, Construction and Evaluation of New Attacks against Multi-modal Mobile GUI Agentscited by 5
  2. 2025Protecting Your Video Content: Disrupting Automated Video-based LLM Annotationscited by 5
  3. 2025Evaluating Robustness of Monocular Depth Estimation with Procedural Scene Perturbationscited by 4
  4. 2024Soft-Label Integration for Robust Toxicity Classificationcited by 1
  5. 2025Dual-Flow: Transferable Multi-Target, Instance-Agnostic Attacks via In-the-wild Cascading Flow Optimizationcited by 1
  6. 2025See No Evil: Adversarial Attacks Against Linguistic-Visual Association in Referring Multi-Object Tracking Systemscited by 1
  7. 2025TestDG: Test-time Domain Generalization for Continual Test-time Adaptationcited by 0
  8. 2025Pay Less Attention to Function Words for Free Robustness of Vision-Language Modelscited by 0
  9. 2026JECA^2: Judgment-Explanation Consistent Adversarial Attack against Forensic Vision-Language Modelscited by 0
  10. 2026On Adversarial Vulnerability of Vision-Language Models through the Lens of Intermediate Spectral Subspacescited by 0
  11. 2026COMIC: Reference-Aware Safety Gating for Multimodal Large Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Grounding-Driven Attack: Improving Encoder-based Adversarial Transferability against Large Vision-Language Models
  2. 2025Transferable Adversarial Attacks on Black-Box Vision-Language Models
  3. 2025IAG: Input-aware Backdoor Attack on VLM-based Visual Grounding
  4. 2026Breaking the weakest link to evade vision language models
  5. 2025Survey of Adversarial Robustness in Multimodal Large Language Models
  6. 2024Towards Adversarially Robust Vision-Language Models: Insights from Design Choices and Prompt Formatting Techniques
  7. 2026Investigating Adversarial Robustness of Multi-modal Large Language Models
  8. 2026FORUM: Frozen Outputs Reconciled Using Model Agreement for Visual Grounding