LATIDIA · Ciberseguridad
PGPO: Ofuscación progresiva guiada por prototipos para recomendaciones mejoradas por LLM que preservan la privacidad
arXiv:2611.23236v1 Tipo de anuncio: cruzado Resumen: Los sistemas de recomendación mejorados con LLM codifican la información lateral de los elementos en incrustaciones semánticas, pero estas incrustaciones pueden filtrar atributos textuales confidenciales a través de ataques de inversión.
WhatsApp ↗Telegram ↗
La noticia
arXiv:2611.23236v1 Announce Type: cross Abstract: LLM-enhanced recommender systems encode item side information into semantic embeddings, but these embeddings may leak sensitive textual attributes through inversion attacks. Existing input-side defenses rely on trusted collection or local token perturbation, often disrupting the inter-item semantic topology needed for recommendation. To address it, we introduce relational fidelity as the key objective, requiring semantic shift for privacy while preserving relative item relations for utility. Based on this principle, we propose prototype-guided progressive obfuscation (PGPO), which learns topology-aware obfuscation anchors for high-influence prototype words and propagates them through a static semantic graph. PGPO is optimized with GRPO using individual dissimilarity, structural consistency, and semantic constraint rewards.