Do membership inference attacks work on large language models?

M Duan, A Suri, N Mireshghallah, S Min, W Shi… - arXiv preprint arXiv …, 2024 - arxiv.org
Membership inference attacks (MIAs) attempt to predict whether a particular datapoint is a
member of a target model's training data. Despite extensive research on traditional machine …

Leveraging Simulation Data to Understand Bias in Predictive Models of Infectious Disease Spread

A Züfle, F Salim, T Anderson, M Scotch… - ACM Transactions on …, 2024 - dl.acm.org
The spread of infectious diseases is a highly complex spatiotemporal process, difficult to
understand, predict, and effectively respond to. Machine learning and artificial intelligence …

AI Risk Management Should Incorporate Both Safety and Security

X Qi, Y Huang, Y Zeng, E Debenedetti… - arXiv preprint arXiv …, 2024 - arxiv.org
The exposure of security vulnerabilities in safety-aligned language models, eg, susceptibility
to adversarial attacks, has shed light on the intricate interplay between AI safety and AI …

Adoption of

R Steed, A Acquisti - … : Drivers, Designs, & Decoupling (February 7 …, 2024 - papers.ssrn.com
Techniques for privacy-preserving analytics (PPA) offer organizations a way to maintain and
expand access to valuable data while preserving individuals' privacy. Adoption of PPA is …

[HTML][HTML] 2024: A Year of Crises, Change, Contemplation, and Commemoration

XL Meng - 2024 - hdsr.mitpress.mit.edu
The editorial for the first January issue of Harvard Data Science Review was titled “2020: A
Very Busy Year for Data Science (and HDSR)”(Meng, 2020). At that time, I envisaged …