#privacy
-
Embedding Inversion: Reconstructing Text From Vectors
Embedding inversion recovers the original text from a model's embedding vectors, breaking the assumption that stored embeddings are privacy-safe.
-
Model Inversion Attacks: Reconstructing Training Data from Output
From Fredrikson's pharmacogenetics exploit to gradient inversion, model inversion attacks recover private training data most ML teams assume is safe.
-
Training Data Extraction from LLMs: The Carlini Results Explained
Carlini et al. demonstrated verbatim extraction of training data from GPT-2. The results have been widely misread. What the extraction numbers actually show.
-
Membership Inference Attacks: What Works on Production ML APIs
Shokri et al.'s shadow-model attack is the canonical reference, but the gap between the paper's threat model and a real rate-limited API is wide.