Be Kind, Rewrite: Benign Projections Via Rewriting Defend against LLM Data Poisoning Attacks

Large language models LLMs are highly susceptible to backdoor attacks BAs, wherein training samples are poisoned using trigger-based harmful content. Furthermore, existing defenses have proven ineffective when extensively tested across BA patterns. To better combat BAs, we explore the use of LLM...

UniAud: a Unified Auditing Framework for High Auditing Power and Utility with One Training Run

Differentially private DP optimization has been widely adopted as a standard approach to provide rigorous privacy guarantees for training datasets. DP auditing verifies whether a model trained with DP optimization satisfies its claimed privacy level by estimating empirical privacy lower bounds...