Unlearnable Examples

Follow this topic Ask about this topic

Unlearnable examples study how to alter training data so models trained on it fail to acquire the intended patterns, offering a way to deter unauthorized use or degrade downstream learning. Methods include adversarial and frequency-based image perturbations, semantic changes, and approaches for text, graphs, video, and multimodal data; many are tested on image-classification benchmarks and against defenses such as augmentation or adversarial training. A central challenge is maintaining protection across pretrained models, data transformations, and adaptive attacks while keeping the content useful and perceptually faithful. Evidence is still strongest in controlled benchmarks, and pretraining priors and real-world processing pipelines can weaken protection or complicate evaluation.

Latest papers

18 papers1 paper in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

A Gradient based Adversarial Augmentation Framework for Mitigating Unlearnable Datasets
Oct 28, 2026Long Dang, Satpal Sura, Kaiqi XiongUnlearnable Examples Data Augmentation
ICNS Lab and Cyber Florida, University of South Florida, FL, USA
KVeil: First-Place Solution to ECCV 2026 LifeGenIP Challenge on Unlearnable Videos
Oct 9, 2026Boheng Li, Yueming Xing, Yiming Li +3Unlearnable Examples Adversarial Perturbations
Nanyang Technological University, Singapore · Wuhan University, China · SiliconProspect AI +1
Leveraging Imperfect Restoration for Data Availability Attack
Sep 4, 2026Yi Huang, Jeremy Styborski, Mingzhi Lyu +2Data Poisoning Attacks Unlearnable Examples
College of Computing and Data Science, Nanyang Technological University, Singapore · Rapid-Rich Object Search (ROSE) Lab, Interdisciplinary Graduate Programme, Nanyang Technological University, Singapore
MUGEN: Generating Unlearnable Graph Examples for Multiple Learning Tasks
Sep 1, 2026Ziyan Liu, Chengshuai Zhao, Huan LiuGraph Neural Networks Graph Representation Learning
Harrisburg University of Science and Technology Harrisburg, Pennsylvania, USA · Arizona State University Tempe, Arizona, USA
Reversible Unlearnable Examples: Towards the Copyright Protection in Deep Learning Era
Aug 6, 2026Binze Wang, Jinyu Tian, Xingrun Wang +2Dataset Watermarking Unlearnable Examples
School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau, China · School of Computer Science and Artificial Intelligence of FoShan University, FoShan 528225, China · Faculty of Applied Sciences of the Macao Polytechnic University, Macau, China
Adversarial Attacks for Good: A Survey of Proactive Protection across the Visual Content Lifecycle
Aug 5, 2026Jiaming Zhang, Boyang Chen, Zherui Li +14Data Provenance Adversarial Robustness
College of Computing and Data Science, Nanyang Technological University, Singapore · School of Computing and Information Systems, University of Melbourne, Australia · School of Computing and Information Systems, Singapore Management University, Singapore +3
TextCloak: Safeguarding Data from Unauthorized LLM Fine-Tuning via RL-Driven Unlearnable Examples
Jul 30, 2026Chengshuai Zhao, Pingchuan Ma, Dawei Li +4Privacy-Preserving Language Models Unlearnable Examples
School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ, USA · Department of Computer Science and Engineering, Texas A&M University, College Station, TX, USA · School of Computing, Stevens Institute of Technology
DiffUE: Enhancing Utility-Unlearnability Trade-off of Unlearnable Examples via Diffusion Autoencoders
Jul 12, 2026Syed Irfan Ali Meerza, Oktay Ozturk, Amir Sadovnik +1Adversarial Examples Privacy-Preserving ML
University of Tennessee, Knoxville, TN, USA · Virginia Commonwealth University, Richmond, VA, USA · Oak Ridge National University, Knoxville, TN, USA +1
Unlearnable Faces: Privacy Protection Surviving Extraction Pipeline
Jul 7, 2026Byunghoon Oh, Sunghwan Park, Jaewoo LeeAdversarial Examples Privacy-Preserving ML
Chung-Ang University, Seoul, South Korea
Full spectrum Unlearnable Examples via Spectral Equalization
Jun 25, 2026Jiale Cai, Gezheng Xu, Zhihao Li +6Adversarial Examples Adversarial Perturbations
Department of Computer Science, Western University, London, ON, Canada · School of Statistics and Data Science, Southeast University, Nanjing, China · Department of Mechanical, Industrial and Aerospace Engineering, Concordia University +2
Toward Understanding Adversarial Distillation: Why Robust Teachers Fail
May 21, 2026Hongsin Lee, Hye Won ChungAdversarial Training Neural Network Robustness
School of Electrical Engineering, KAIST, Daejeon, Korea.
VoxShield: Protecting 3D Medical Datasets from Unauthorized Training via Frequency-Aware Inter-Slice Disruption
May 17, 2026Xinyao Liu, Zhipeng Deng, Wenhan Jiang +4Adversarial Examples 3D Medical Image Segmentation
Westlake University, Hangzhou, China · Dalian University of Technology, Dalian, China · Hokkaido University, Sapporo, Japan +2
The Unlearnability Phenomenon in RLVR for Language Models
May 16, 2026Yulin Chen, He He, Chen ZhaoReinforcement Learning RL for Language Model Reasoning
1New York University · 2NYU Shanghai
To See is Not to Learn: Protecting Multimodal Data from Unauthorized Fine-Tuning of Large Vision-Language Model
May 14, 2026Chengshuai Zhao, Zhen Tan, Dawei Li +2VLM Robustness Multimodal Robustness
School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ, USA · Department of Computer Science and Engineering, Texas A&M University, College Station, TX, USA.
Dual-branch Robust Unlearnable Examples
May 3, 2026Xianlong Wang, Hangtao Zhang, Wenbo Pan +4Adversarial Robustness Adversarial Examples
Department of Computer Science, City University of Hong Kong, Hong Kong SAR, China · University of Pennsylvania, Philadelphia, USA · Huazhong University of Science and Technology, Wuhan, China +2
Channel-Level Semantic Perturbations: Unlearnable Examples for Diverse Training Paradigms
Apr 18, 2026Bo Wang, Jia Ni, Mengnan Zhao +2Adversarial Examples Privacy-Preserving ML
School of Information and Communication Engineering, Dalian University of Technology, Dalian 116024, China · School of Computer Science and Technology, Anhui University, Hefei 230601, China · School of Computer Science and Technology, Zhejiang University, Hangzhou 310058, China
When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretraining
Mar 5, 2026Zhihao Li, Gezheng Xu, Jiale Cai +5Unlearnable Examples Adversarial Perturbations
Western University · Concordia University · Beijing University of Posts and Telecommunications +1
Why Do Unlearnable Examples Work: A Novel Perspective of Mutual Information
Mar 4, 2026Yifan Zhu, Yibo Miao, Yinpeng Dong +1Unlearnable Examples Privacy-Preserving ML
State Key Laboratory of Mathematical Sciences, Academy of Mathematics and Systems Science, Chinese Academy of Sciences · University of Chinese Academy of Sciences · College of AI, Tsinghua University +1

Oct 28, 2026J/K move · Enter open · S save

A Gradient based Adversarial Augmentation Framework for Mitigating Unlearnable Datasets

Long Dang, Satpal Sura, Kaiqi Xiong

ICNS Lab and Cyber Florida, University of South Florida, FL, USA

Unlearnable Examples Data Augmentation

Open paper AskPDF arXiv

Unlearnable data methods (ULD) protect public data from unauthorized model training by adding imperceptible perturbations that create a learning shortcut. A deep neural network can exploit the shortcut learning path to minimize training loss without learning generalizable features. Intuitively, finding adversarial versions of perturbed images that can substantially increase the model's loss can force the model to learn generalizable features. Existing data augmentation based attacks rely on heuristic search strategies/random sampling over a curated augmentation set to identify adversarial augmentation operations, an approach that often does not scale when the number of candidate operations is large. To address the challenge of a large search space, we propose {\techname}, a gradient based adversarial augmentation framework that learns an augmentation policy via stochastic gradient ascent. {\techname} leverages a differentiable data augmentation pipeline to efficiently search a set of up to 18 augmentation techniques, from which a grid search selects a protection-specific subset. The augmentation subset is trained to maximize a surrogate model's classification loss. We introduce a perceptual-plus-pixel regularization to learn operation parameters that preserve image semantics. Under a controlled ablation, {\techname} consistently outperforms an equivalent random sampling baseline (80.0% vs. 77.29% clean test accuracy on EMIN-protected CIFAR-10 for ResNet-18 and 83.28% vs. 66.45% on NTGA-protected CIFAR-10 for MobileNet-V2), demonstrating that gradient guidance drives the clean test accuracy recoverability rather than the expanded search space. {\techname} offers an effective and scalable adversarial augmentation framework for stress-testing ULD methods, a critical tool for developing robust ULD methods.