Open access
2026
A Multi-Source Retrieval-Augmented Large Language Model Architecture for Trustworthy Conversational AI in Diabetes Care
The proposed Retrieval-Augmented Large Language Model architecture for trustworthy conversational AI in diabetes care is presented, which combines clinical entity recognition, hybrid dense–sparse retrieval, patient-context filtering, cross-encoder reranking, evidence-constrained prompt construction, and multi-layer safety verification to generate personalized, evidence-supported conversational responses.
Muhammad Jamil, A. Kavak, Sema Bayraktar et al.
· IEEE Access · 0 citations