Adversarial Examples

Ask about this topic

Overview

This field studies how small or structured changes to inputs can exploit model vulnerabilities, which matters for the reliability and safety of deployed AI. Attacks range from gradient-based pixel perturbations and patches to black-box natural transformations, latent-space methods, and prompt-based attacks on vision-language and generative systems. Defenses include adversarial training, input filtering and detection, robust representation learning, and protective perturbations for images, audio, and 3D data. Robustness remains difficult to measure: attacks can transfer across models or survive physical pipelines, while gains on standard benchmarks may fail against adaptive, semantic, or multimodal attacks and trade off against utility or fidelity.

Latest papers

CardsList

76 papers5 papers in the last four weeks, up 25% on the four weeks before. 0.0% of all new papers.

  1. LETHE: Defending against Unauthorized Image-to-Video Generation via Value Diversity Collapse

    Oct 9, 2026Boheng Li, Yueming Xing, Yiming Li +2Image-to-Video GenerationAdversarial Examples

    Nanyang Technological University · Wuhan University · SiliconProspect AI

  2. Collaboratively Guided Adversarial Robust Distillation with Teacher-Favorable Examples

    Oct 8, 2026Zhi Li, Haowei Liu, Hongchen Yang +4Adversarial RobustnessAdversarial Knowledge Distillation

    Engineering Research Center of Cyberspace and School of Software and AI, Yunnan University, Kunming, China · School of Information Science and Technology, Yunnan Normal University, Kunming, China

  3. SPIN: Image Immunization Against Diffusion Editing via Single-Step Projection in Stochastic Neighborhoods

    Oct 5, 2026Fengming Gu, Jie Zhang, Zhongqi Wang +3Adversarial ExamplesDiffusion-Based Image Editing

    School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences · State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences · University of Chinese Academy of Sciences +1

  4. Anti-Persona: Disrupting Unauthorized Identity Binding and Recognition in Personalized Vision--Language Models

    Oct 1, 2026Abhishek Basu, Fahad Shamshad, Karthik NandakumarVLM RobustnessVision-Language Models

    MBZUAI, UAE · Michigan State University, USA

  5. CollageAttack: Exploiting Cross-Modal Alignment Flaws in T2I Models through Spatial Text Composition

    Sep 29, 2026Zhiyi Mou, Yao Lu, Wangze Ni +6Adversarial ExamplesVision-Language Alignment

    Zhejiang University · Hong Kong Polytechnic University

  6. Evaluating the Semantic-to-Geometric Gap in Adversarial Defenses Against Vision-Language Model-Based Plagiarism

    Sep 22, 2026Christopher Burger, Christina Trotter, Joseph Carlisle +1VLM RobustnessAdversarial Attacks on VLMs

    The University of Mississippi

  7. Discovering Natural Transformation Vulnerabilities in Black-Box Vision Models

    Sep 7, 2026Dongsu Song, DaeYun GO, Jay Hoon JungAdversarial AttacksAdversarial Examples

  8. Generating Adversarial Texts for Machine Translation via GRPO

    Sep 5, 2026Florian Zogaj, Jakob Hütteneder, Giovanni De Muri +3Adversarial Examples

    ETH Zurich, Switzerland

  9. Audio-Driven Adversarial Defense for 3D Talking Face Generation with totally Visual Fidelity Preservation

    Aug 31, 2026Rui-Qing Sun, Chen-Hao Cui, Hui-Yang Zhao +3Audio-Driven Facial AnimationAdversarial Examples

    Beijing Institute of Technology Haidian Qu, Beijing Shi, China

  10. Adversarial Training Without Input Gradients via Low-Rank Householder Expansions

    Aug 27, 2026Tiana C. Johnson, Donsub RimAdversarial TrainingAdversarial Robustness

  11. Multi2AV-Safety: Benchmarking Safety in Multimodal-to-Audio-Video Generation

    Aug 27, 2026Kaichao Jiang, Changtao Miao, Baiqi Wu +8Adversarial Attacks on Unified Multimodal ModelsAudio-Video Generation

    University of Science and Technology of China · Independent Researcher · Zhejiang University +1

  12. Why Does Robustness Reduce Superposition?

    Aug 23, 2026Adam ElimadiFeature SuperpositionAdversarial Examples

    Independent Researcher

  13. GuardPaint:SpeculativeSafetyDecodingforText-to-ImageGeneration

    Aug 22, 2026Shreyash Dhoot, Paras Dhiman, Arsh Abbas Naqvi +4Safe T2I GenerationAdversarial Attacks on Diffusion Models

    Pragya Lab, BITS Pilani Goa, India · Apple, USA · Meta, USA

  14. ArmorOCR: Grounded Adversarial Visual Perception via Observation-Transferred Self-Distillation

    Aug 20, 2026Linhan Cao, Siyuan Li, Jun Lan +8Adversarial RobustnessAdversarial Examples

    Ant Group · Shanghai Jiao Tong University · East China Normal University

  15. Training-Free Reconstruction-Based AI-Generated Image Detectors Are Inherently Vulnerable to Adversarial Examples

    Aug 17, 2026Roman Demchenko, Jonas Ricker, Asja FischerAdversarial Attacks on AI-Generated Image DetectorsAI-Generated Image Detection

    Ruhr University Bochum, Bochum, Germany

  16. Augmenting Text to Increase Translation Difficulty

    Aug 16, 2026William Kalikman, Šimon Sukup, Michal Tešnar +1MT EvaluationAdversarial Examples

    ETH Zurich

  17. SPARED: Reasoning-Based AI-Generated Image Detection via Adversarially Edited Data

    Aug 13, 2026Yicheng Bao, Xiahui Guo, Xuhong Wang +1Adversarial TrainingAdversarial Examples

    1East China Normal University · 2Shanghai Artificial Intelligence Laboratory

  18. SRAP: SVD-Refined Adversarial Perturbations for Imperceptible Face-Swap Defense

    Aug 4, 2026Sungwon Cho, Kwanghyun Ko, Myungjoo KangAdversarial ExamplesAdversarial Perturbations

    Interdisciplinary Program in Artificial Intelligence Seoul National University · Interdisciplinary Program in Computational Science and Technology Seoul National University · Department of Mathematical Sciences Seoul National University

  19. VETO: Towards Protecting Images From Frontier AI Editing

    Jul 29, 2026Jonas Grebe, Hossein Shakibania, Tobias Braun +2Image Editing EvaluationAdversarial Examples

    TU Darmstadt & hessian.AI, Germany · Zuse School ELIZA

  20. InkShield: Writing Style Protection Against Unauthorized Handwriting Mimicry

    Jul 29, 2026Jian Xiong, Wenbo Jiang, Zihan Wang +4Adversarial Examples

    University of Electronic Science and Technology of China

  21. I2VShield: An Efficient Proactive Defense Framework against DiT-based Image-to-Video Models

    Jul 28, 2026Yimao Guo, Zuomin Qu, Wei LuImage-to-Video GenerationAdversarial Attacks

    School of Computer Science and Engineering, Sun Yat-sen University Ministry of Education Key Laboratory of Information Technology Guangdong Province Key Laboratory of Information Security Technology Guangzhou 510006, China · Guangdong Province Key Laboratory of Information Security Technology Guangzhou 510006, China · State Key Laboratory of HVDC, China Southern Power Grid Electric Power Research Institute

  22. ISPCloak: Weaponizing ISP for Optimization-Free Physical Camouflage against Deepfake Detectors

    Jul 24, 2026Jiale Zhao, Jiajun Wan, Lei Tang +3Adversarial AttacksDeepfake Detection

    Guangdong University of Technology Guangzhou, China · Guizhou Provincial Laboratory of Big Data, State Key Laboratory of Public Big Data, Guizhou University Guiyang, China

  23. A New Kind of Adversarial Example: Measuring the Human-Model Gap, and Its Relationship to OOD Detection

    Jul 22, 2026Ali BorjiAdversarial RobustnessAdversarial Examples

    BrainChip Inc., Laguna Hills, CA, USA

  24. Dynamic Defense Profiling Enables Cognitive Jailbreak of Text-to-Image Models

    Jul 20, 2026Dongdong Yang, Deyue Zhang, Zhao Liu +5Adversarial Prompt GenerationAdversarial Examples

    AI Security Lab, Beijing, China · School of Artificial Intelligence, Beihang University, Beijing, China

  25. Semantic Color Naturalness Breaker: Preventing Illegitimate Colorization via Content-Aware Color Priors

    Jul 20, 2026Yuki Nii, Futa Waseda, Ching-Chun Chang +1Adversarial Examples

    The University of Tokyo, Japan · National Institute of Informatics, Japan

  26. Automatic Hard Example Synthesis with Multi-Level Agentic Data Curation

    Jul 15, 2026Genglin Liu, Muye Zhang, Krishnamurthy Viswanathan +5Multimodal RobustnessAdversarial Scenario Generation

    1UCLA · *Work done as a student researcher at Google. · 2Google

  27. DiffUE: Enhancing Utility-Unlearnability Trade-off of Unlearnable Examples via Diffusion Autoencoders

    Jul 12, 2026Syed Irfan Ali Meerza, Oktay Ozturk, Amir Sadovnik +1Adversarial ExamplesPrivacy-Preserving ML

    University of Tennessee, Knoxville, TN, USA · Virginia Commonwealth University, Richmond, VA, USA · Oak Ridge National University, Knoxville, TN, USA +1

  28. Imperceptible and Reversible Adversarial Examples against Vision-Language Models for Privacy Protection

    Jul 11, 2026Qi Lu, Ziqi Zhou, Yufei Song +5VLM RobustnessAdversarial Attacks on VLMs

    School of Cyber Science and Engineering, Huazhong University of Science and Technology · College of Computer Science, Chongqing University · School of Software and engineering, Huazhong University of Science and Technology +2

  29. Statistically Undetectable Backdoors in Deep Neural Networks

    Jul 10, 2026Andrej Bogdanov, Alon Rosen, Neekon VafaBackdoor AttacksAdversarial Examples

    University of Ottawa. · Bocconi University. · Massachusetts Institute of Technology.

  30. Adversarially Guided Diffusion for LiDAR Range Image Synthesis

    Jul 8, 2026Stavros Bouras, Antonios Makris, Alexandros Gkillas +2Adversarial AttacksAdversarial Examples

    School of Electrical and Computer Engineering, National Technical University of Athens, Greece · Industrial Systems Institute, Athena Research Center, Patras Science Park, Greece