5 accepted papers
Large language models (LLMs) can be prompted with specific styles (e.g., formatting responses as lists), including in malicious queries. Prior jailbreak research mainly augments these queries with additional string transformations to maximize attack success rate (ASR). However, the impact of style p…
For an LLM to correctly respond to an instruction it must understand both the semantics and the domain (i.e., subject area) of a given task-instruction pair. However, syntax can also convey implicit information. Recent work shows that \textit{syntactic templates}---frequent sequences of Part-of-Spee…
Privacy estimation techniques for differentially private (DP) algorithms are useful for comparing against analytical bounds, or to empirically measure privacy loss in settings where known analytical bounds are not tight. However, existing privacy auditing techniques usually make strong assumptions o…
Machine learning models are often personalized with categorical attributes that define groups. In this work, we show that personalization with *group attributes* can inadvertently reduce performance at a *group level* -- i.e., groups may receive unnecessarily inaccurate predictions by sharing their…
We study the problem of deleting user data from machine learning models trained using empirical risk minimization (ERM). Our focus is on learning algorithms which return the empirical risk minimizer and approximate unlearning algorithms that comply with deletion requests that come in an online manne…