# VALD: Multi-Stage Vision Attack Detection for Efficient LVLM Defense

arXiv: https://arxiv.org/abs/2602.19570 · HTML: https://mlsanity.com/paper/2602.19570/

## Key facts

- Cited by 1 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 21 of its 38 references that are in the map.
- First posted on arXiv 23 Feb 2026.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM jailbreaks and adversarial robustness](https://mlsanity.com/mllm/mllm-jailbreaks-and-adversarial-robustness/)

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [On Evaluating Adversarial Robustness](https://mlsanity.com/paper/1902.06705/) (2019)
- [Visual Adversarial Examples Jailbreak Aligned Large Language Models](https://mlsanity.com/paper/2306.13213/) (2023)
- [Image Hijacks: Adversarial Images can Control Generative Models at Runtime](https://mlsanity.com/paper/2309.00236/) (2023)
- [Eyes Closed, Safety On: Protecting Multimodal LLMs via Image-to-Text Transformation](https://mlsanity.com/paper/2403.09572/) (2024)
- [MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance](https://mlsanity.com/paper/2401.02906/) (2024)
- [How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs](https://mlsanity.com/paper/2311.16101/) (2023)
- [Natural Language Understanding and Inference with MLLM in Visual Question Answering: A Survey](https://mlsanity.com/paper/2411.17558/) (2024)
- [Fight Back Against Jailbreaking via Prompt Adversarial Tuning](https://mlsanity.com/paper/2402.06255/) (2024)
- [A Survey of Safety on Large Vision-Language Models: Attacks, Defenses and Evaluations](https://mlsanity.com/paper/2502.14881/) (2025)
- [ETA: Evaluating Then Aligning Safety of Vision Language Models at Inference Time](https://mlsanity.com/paper/2410.06625/) (2024)
- [JailGuard: A Universal Detection Framework for LLM Prompt-based Attacks](https://mlsanity.com/paper/2312.10766/) (2023)
- [Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks](https://mlsanity.com/paper/2411.15720/) (2024)
- [A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1](https://mlsanity.com/paper/2503.10635/) (2025)
- [Safeguarding Vision-Language Models Against Patched Visual Prompt Injectors](https://mlsanity.com/paper/2405.10529/) (2024)
- [Visual question answering: from early developments to recent advances -- a survey](https://mlsanity.com/paper/2501.03939/) (2025)
- [PIP: Detecting Adversarial Examples in Large Vision-Language Models via Attention Patterns of Irrelevant Probe Questions](https://mlsanity.com/paper/2409.05076/) (2024)
- [Defending LVLMs Against Vision Attacks through Partial-Perception Supervision](https://mlsanity.com/paper/2412.12722/) (2024)

## Built on by (most cited first, 1 of 1)

- [Structure-Guided Visual Perturbation Neutralization for LVLMs](https://mlsanity.com/paper/2605.27927/) (2026)

## Abstract (arXiv)

Large Vision-Language Models (LVLMs) can be vulnerable to adversarial images that subtly bias their outputs toward plausible yet incorrect responses. We introduce a general, efficient, and training-free defense that combines image transformations with agentic data consolidation to recover correct model behavior. A key component of our approach is a two-stage detection mechanism that quickly filters out the majority of clean inputs. We first assess image consistency under content-preserving transformations at negligible computational cost. For more challenging cases, we examine discrepancies in a text-embedding space. Only when necessary do we invoke a powerful LLM to resolve attack-induced divergences. A key idea is to consolidate multiple responses, leveraging both their similarities and their differences. We show that our method achieves state-of-the-art accuracy while maintaining notable efficiency: most clean images skip costly processing, and even in the presence of numerous adversarial examples, the overhead remains minimal.