1 results for tag "ctf-ai-ml"
Quick-reference cheat sheet for AI/ML CTF challenges covering model weight perturbation negation, model inversion, encoder collisions, LoRA adapter merging/exploitation, model extraction and membership inference, adversarial examples (FGSM/PGD/C&W), adversarial patches, data poisoning, and LLM attacks (prompt injection, jailbreaking, token smuggling, tool-use exploitation). Bundles ready-to-run inspection commands for `.pt`/`safetensors`/HuggingFace/LoRA models and points to detail files (`model-attacks.md`, `adversarial-ml.md`, `llm-attacks.md`) for each technique.