Model
FuseRAG
A reasoning acceleration method for GraphRAG that separates reusable content from query-dependent structural interactions and injects structural information directly into the decoder's self-attention layers, achieving lossless TTFT acceleration.
Timeline (last 90 days)
Oct 19
FuseRAG achieves a 2.2 to 2.7 times TTFT speedup over the best GraphRAG baseline with no quality loss by separating reusable content from query-relevant structural interactions and injecting the latter into the decoder's self-attention layers.
Unverified50%