Privacy-Preserving Prompt Personalization in Federated Learning for Multimodal Large Language Models

🗓️ 28 May 2025 00:00:00Reported by Baturalp Buyukates, Songze Li, Sizai HouType

packetstormnews

packetstormnews🔗 packetstorm.news👁 4 Views

SecFPP enables secure federated prompt personalization for multimodal language models using hierarchical prompts and secret sharing.

28 May 2025 00:00Current

6.7Medium risk

Vulners AI Score6.7

federated learning privacy secret sharing adaptive clustering domain adaptation class adaptation hierarchical prompts prompt personalization data heterogeneity multimodal models