Preprint
Oct 2026
FuseRAG: Fusing Structure and Evidence into Inference for Cost-Efficient Graph-based RAG
This work proposes FuseRAG, a cost-efficient graph-based RAG framework that shifts most retrieved evidence from explicit textual contexts to latent representations, and preserves graph structure through latent interactions, enables evidence reuse beyond exact token-prefix matching.
Ting-Qi Wang, Ying-Shu Li, Zhi-Peng Cai
· 0 citations