AI RESEARCH
MAVEN: Multi-Agent Verification-Elaboration Network with In-Step Epistemic Auditing
arXiv CS.AI
•
ArXi:2605.07646v1 Announce Type: cross While explicit reasoning trajectories enhance model interpretability, existing paradigms often rely on monolithic chains that lack intermediate verification, allowing early errors to cascade unchecked. This lack of modularity impedes granular auditing and compromises the epistemic trust required for high-stakes applications. We propose MAVEN (Multi-Agent Verification-Elaboration Network with In-Step Epistemic Auditing), a blackboard-inspired framework designed to transform LLMs into deliberate reasoners through explicit role-decoupling.