← Research
Accepted PaperAccepted · ACL TrustNLP 2026Accepted · SAGAI 2026 · IEEE S&P colocated

Authorization-First Retrieval: Enforcing Least Privilege in Multi-Agent RAG Systems

· Rohith Namboothiri · ACL TrustNLP 2026; SAGAI 2026, colocated with IEEE S&P 2026
Read full paper

Abstract

Authorization-first retrieval enforces least privilege before semantic retrieval, ensuring that documents never enter a multi-agent RAG prompt unless the requesting user or agent is permitted to access them.

Venue
ACL TrustNLP 2026; SAGAI 2026, colocated with IEEE S&P 2026
DOI
10.36227/techrxiv.177273889.98246164/v1

Generation-time authorization assumes the model can be trusted to refuse what it has already seen. Authorization-first retrieval inverts the contract: the document never enters the prompt unless the requesting agent has explicit permission for it.

The paper formalizes the threat model for multi-agent RAG, presents implementation patterns compatible with vector and hybrid retrieval, and argues that authorization must constrain the candidate set before ranking or synthesis.

Accepted at ACL TrustNLP 2026. Version 1 was also accepted at SAGAI 2026, colocated with IEEE Symposium on Security and Privacy 2026.

Keywords

RAGAuthorizationMulti-agentSecurityRBAC