3 accepted papers
Large language models can memorize information that must be removed--ranging from copyright-sensitive content (e.g., book chapters) to personally identifiable information (e.g., income)--to ensure responsible and compliant behavior. Unlearning has emerged as an efficient alternative to full retraini…
Text-to-image (T2I) models are increasingly popular, producing a large share of AI-generated images online. To compare model quality, voting-based leaderboards have become the standard, relying on anonymized model outputs for fairness. In this work, we show that such anonymity can be easily broken.
We introduce One-Shot Label-Only (OSLO) membership inference attacks (MIAs), which accurately infer a given sample's membership in a target model's training set with high precision using just a single query, where the target model only returns the predicted hard label. This is in contrast to stat…