From the 1 of 4 linked papers with an AI index.
6 papers
Extractable Memorization From First Principles
A. Feder Cooper, Marika Swanberg, Jamie Hayes +5
The paper introduces formal matched‑comparison methods—using conformal testing and document‑level censuses—to reliably determine when a language model has memorized training data,…
A Unified Framework for Adversary-Aware Differential Privacy Bounds
Marika Swanberg, Meenatchi Sundaram Muthu Selva Annamalai, Jamie Hayes +2
Differential Privacy (DP) bounds the privacy leakage of a mechanism against worst-case membership inference, but the precise tradeoff between complex adversarial models and DP prot…
Measuring memorization in language models via probabilistic extraction
Jamie Hayes, Marika Swanberg, Harsh Chaudhari +6
Large language models (LLMs) are susceptible to memorizing training data, raising concerns about the potential extraction of sensitive information at generation time. Discoverable…
Privacy in Metalearning and Multitask Learning: Modeling and Separations
Maryam Aliakbarpour, Konstantina Bairaktari, Adam Smith +2
Model personalization allows a set of individuals, each facing a different learning task, to train models that are more accurate for each person than those they could develop indiv…
Auditing Privacy Mechanisms via Label Inference Attacks
Róbert István Busa-Fekete, Travis Dick, Claudio Gentile +3
We propose reconstruction advantage measures to audit label privatization mechanisms. A reconstruction advantage measure quantifies the increase in an attacker's ability to infer t…
ATTAXONOMY: Unpacking Differential Privacy Guarantees Against Practical Adversaries
Rachel Cummings, Shlomi Hod, Jayshree Sarathy +1
Differential Privacy (DP) is a mathematical framework that is increasingly deployed to mitigate privacy risks associated with machine learning and statistical analyses. Despite the…