4 papers
Fragments to Facts: Partial-Information Fragment Inference from LLMs
Lucas Rosenblatt, Bin Han, Robert Wolfe +1
Large language models (LLMs) can leak sensitive training data through memorization and membership inference attacks. Prior work has primarily focused on strong adversarial assumpti…
Are Data Experts Buying into Differentially Private Synthetic Data? Gathering Community Perspectives
Lucas Rosenblatt, Bill Howe, Julia Stoyanovich
Data privacy is a core tenet of responsible computing, and in the United States, differential privacy (DP) is the dominant technical operationalization of privacy-preserving data a…
ML-EAT: A Multilevel Embedding Association Test for Interpretable and Transparent Social Science
Robert Wolfe, Alexis Hiniker, Bill Howe
This research introduces the Multilevel Embedding Association Test (ML-EAT), a method designed for interpretable and transparent measurement of intrinsic bias in language technolog…
Dataset Scale and Societal Consistency Mediate Facial Impression Bias in Vision-Language AI
Robert Wolfe, Aayushi Dangol, Alexis Hiniker +1
Multimodal AI models capable of associating images and text hold promise for numerous domains, ranging from automated image captioning to accessibility applications for blind and l…