LATIDIA · Ciberseguridad

SnapAudit: Auditoría activa del aprendizaje en contexto diferencialmente privado a través de la simulación basada en instantáneas

arXiv:2511.13502v4 Announce Type: replace Resumen: El aprendizaje en contexto (ICL) permite a los LLM adaptarse a las tareas utilizando demostraciones que pueden contener datos confidenciales. Los mecanismos de ICL diferencialmente privados (DP) mitigan esto

La noticia

arXiv:2511.13502v4 Announce Type: replace Abstract: In-context learning (ICL) enables LLMs to adapt to tasks using demonstrations that may contain sensitive data. Differentially private (DP) ICL mechanisms mitigate this risk by adding noise during aggregation. However, verifying their claimed privacy bounds through repeated end-to-end membership-inference attacks is costly and yields unstable estimates. We propose SnapAudit, an active auditing framework that decomposes a DP-ICL pipeline into a deterministic clean-inference stage and a stochastic DP-noise stage, and audits the full pipeline by combining a small snapshot of the former with bootstrap simulation of the latter. Because clean LLM outputs are near-deterministic at low temperature, tens of thousands of clean LLM calls suffice to