Every classifier has a blind spot.
Adversarial ML coverage for engineers shipping ML systems. Evasion, data poisoning and backdoors, model extraction, membership inference, training-data extraction, and the multimodal attacks that ride a shared vision encoder into every model downstream of it. Built from the published papers and reference implementations, with the threat model each result assumes stated up front.
Enter the archive →Latest entries
Backdoor Attack on Neural Networks Detection: What Works
defensesHow Adversarial Examples Fool Image Classifiers
primerFraise Agent Memory Is a Stored Prompt-Injection Surface
attacksPreprocessing Defenses Fail on Depthwise-Separable Edge CNNs
defensesHow FGSM Adversarial Attacks Work: Fast Gradient Sign Method
primerCLIP Adversarial Attacks: One Perturbation, Every Model
attacksAdversarial Attack Libraries: ART, Foolbox, torchattacks
toolingGraphRP Defense Against GNN Model Extraction
defensesAdversarial ML Testing Tools: ART, ModelScan, Giskard
toolingData Poisoning Attack Detection Methods That Actually Work
defensesStart here: one entry per attack family
The archive runs newest first. This is the other way in — the reference piece for each family, from the perturbation that fools one classifier to the one perturbation that steers every model built on a shared encoder.
- Evasion Evasion Attacks on Image Classifiers: FGSM, PGD, and C&W
- Multimodal Adversarial Attacks on Vision-Language Models
- CLIP CLIP Adversarial Attacks: One Perturbation, Every Model
- Poisoning Data Poisoning and Backdoor Attacks on Foundation Models
- Privacy Membership Inference Attacks on Production ML APIs
- Physical Adversarial Patch Attacks in the Physical World
- Evaluation Evaluating Adversarial Robustness Without Fooling Yourself
- Tooling Adversarial Attack Libraries: ART, Foolbox, torchattacks
- Black-box Adversarial Transferability: How Black-Box Attacks Work
- Certified Certified Robustness via Randomized Smoothing
- Text Adversarial Robustness in NLP: Why Text Attacks Differ
- Federated Gradient Inversion Attacks in Federated Learning
- LLM privacy Training Data Extraction from LLMs: Attacks and Defenses
Not sure which apply to your model? Filter the attack catalogue by access level, modality, adversary goal and the defenses already in place.
Independent, specialist, and free to read
Adversarial ML publishes focused, sourced guides on a single topic. No paywall, no account, no ad tracking.
Adversarial ML — in your inbox
Working adversarial ML — exploits, defenses, and the gap between — delivered when there's something worth your inbox.
No spam. Unsubscribe anytime.