AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

PhysPatch: A Physically Realizable and Transferable Adversarial Patch Attack for Multimodal Large Language Models-based Autonomous Driving Systems

Qi Guo, Xiaojun Jia, Shanmin Pang, Simeng Qin, Lin Wang, Ju Jia, Yang Liu, Qing Guo

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Vision-language model robustness · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add potential-field-based mask update instead of fixed 120x120 square patch (ablation)PhysPatch · nuScenes stop-sign targetASR (%): 37.9 -> 40.3 (GPT-4o)Improved
add SVD-based local alignment loss instead of standard local alignment (ablation)PhysPatch · nuScenes stop-sign targetASR (%): 38.8 -> 40.3 (GPT-4o)Improved
add patch-guided crop-resize instead of random cropping (ablation)PhysPatch · nuScenes stop-sign targetASR (%): 37.5 -> 40.3 (GPT-4o)Improved
add SVD-based local alignment loss (ablation)PhysPatch · nuScenes stop-sign targetASR (%): 14.0 -> 14.5 (Claude-4)Improved
add potential-field mask update (ablation)PhysPatch · nuScenes stop-sign targetASR (%): 22.9 -> 25.4 (Gemini-2.5)Improved
use PhysPatch instead of FOA-Attack (patch of ~1% image area)MLLM-based AD systems · nuScenes stop-sign, QVQ-PlusASR (%): 20.5 -> 29.2Improved
use PhysPatch instead of FOA-Attack under PAD defenseGPT-4o · nuScenes stop-sign, PAD defenseASR (%): 6.7 -> 16.8Improved
PhysPatch vs FOA-Attack on Qwen2.5-VL-72BQwen2.5-VL-72B · nuScenes stop-sign targetASR (%): 14.4 -> 15.4No clear effect
PhysPatch vs FOA-Attack on generation timeattack optimization · Generation timeseconds: 174 -> 152Improved
PhysPatch vs lightweight baselines on generation timeattack optimization · Generation timeseconds: 79 -> 152 (AttackVLM)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy