LATIDIA · Ciberseguridad

TranScope: Lo que el software oculta sobre los datos de capacitación de LLM, el hardware revela a escala y los aceleradores se magnifican

arXiv: 2610.06848v1Tipo de anuncio: nuevo Resumen: La membresía es la primitiva de privacidad raíz en el aprendizaje automático: hasta la fecha, no se ha demostrado ninguna detección de fuera de distribución basada en hardware en modelos de caja negra contra con

La noticia

arXiv:2610.06848v1 Announce Type: new Abstract: Membership is the root privacy primitive in machine learning: to date, no hardware-based out-of-distribution detection on black-box models has been demonstrated against constant-time, static neural networks with masked confidence. This paper performs the first cycle-level examination of how large language models and vision transformers interact with various modern microarchitecture components, including integrated accelerators, as LLMs scale in size and answers the question of whether the data that a model was trained on affects its execution footprint even without any input-dependent branch, dynamic optimization, or early exit and in constant-time models. The results confirm that the answer is yes and identify which modern hardware components, such