SOURCE-LINKED INTELLIGENCE

VikingRAG: Accurate and Token-efficient Retrieval-augmented Generation over Structured Documents

arXiv · AI, language, vision and robotics · article · Sep 10, 2026 · UTC

State-of-the-art retrieval-augmented generation (RAG) methods exploit document structures to acquire sufficient evidence, but often incur substantial token costs. To reduce structural-context tokens without compromising high RAG accuracy, we present {\sf VikingRAG}, a directory-aware semantic data management system that tightly integrates semantic and structural access to support structural-context-efficient, evidence-gap-driven multi-round retrieval. To further reduce token overhead of multi-round interaction, we materialize agentic multi-round retrieval traces as experience edges, and reuse

Read original source ↗ Open in workspace

recordType

paper

region

Global