LATIDIA · Ciberseguridad
HE-LRM: Modelos de recomendación de aprendizaje profundo cifrados mediante cifrado totalmente homomórfico
arXiv:2506.18150v5 Tipo de anuncio: sustitución Resumen: Los sistemas de inferencia FHE se centran principalmente en modelos con entradas densas, mientras que los modelos de recomendación de aprendizaje profundo (DLRM) contienen características categóricas dispersas que requieren búsquedas.
WhatsApp ↗Telegram ↗
La noticia
arXiv:2506.18150v5 Announce Type: replace Abstract: FHE inference systems primarily target models with dense inputs, while Deep Learning Recommendation Models (DLRMs) contain sparse categorical features that require lookups into embedding tables. Encrypted lookups are challenging because indexing is not natively supported and must be expressed using SIMD arithmetic. A direct one-hot formulation does not scale. Prior work addresses parts of this problem through embedding compression or compact encrypted queries. However, it does not address complete DLRM inference, leaving open how private lookup interacts with downstream computation and memory constraints. We present HE-LRM, the first end-to-end FHE system for private DLRM inference. We use deterministic compositional embedding compression with client-side digit decomposition,