SCOUT-RAG: Scalable and Cost-Efficient Unifying Traversal for Agentic Graph-RAG over Distributed Domains

AAAI 2026 Workshop TrustAgent Submission75 Authors

Keywords: RAG; Agentic AI; Graph-Based Retrieval; Multi-Domain Reasoning; Privacy

TL;DR: This paper introduces SCOUT-RAG (Scalable COst-efficient Unifying Traversal), a distributed agentic Graph-RAG framework that performs progressive, cross-domain retrieval guided by incremental utility signals.

Abstract:

Graph-RAG improves LLM reasoning using structured knowledge, yet conventional designs rely on a centralized knowledge graph. In distributed and access-restricted settings (e.g., hospitals or multinational organizations), retrieval must select relevant domains and appropriate traversal depth without global graph visibility or exhaustive querying.
To address this challenge, we introduce $\textbf{SCOUT-RAG}$ ($\textit{$\underline{S}$calable and $\underline{CO}$st-efficient $\underline{U}$nifying $\underline{T}$raversal}$), a distributed agentic Graph-RAG framework that performs progressive cross-domain retrieval guided by incremental utility goals. SCOUT-RAG employs four cooperative agents that: (i) estimate domain relevance, (ii) decide when to expand retrieval to additional domains, (iii) adapt traversal depth to avoid unnecessary graph exploration, and (iv) synthesize the high-quality answers. The framework is designed to minimize retrieval regret, defined as missing useful domain information, while controlling latency and API cost. Across multi-domain knowledge settings, SCOUT-RAG achieves performance comparable to centralized baselines, including DRIFT and exhaustive domain traversal, while substantially reducing cross-domain calls, total tokens processed, and latency.

Submission Number: 75