Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Xing, Cui, Yanwei, Wang, Guanghui, Qiu, Wei, Li, Ziyuan, Han, Fangwei, Huang, Yajing, Qiu, Hengzhi, Zhu, Bing, He, Peiyang
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915863890755584
author Zhang, Xing
Cui, Yanwei
Wang, Guanghui
Qiu, Wei
Li, Ziyuan
Han, Fangwei
Huang, Yajing
Qiu, Hengzhi
Zhu, Bing
He, Peiyang
author_facet Zhang, Xing
Cui, Yanwei
Wang, Guanghui
Qiu, Wei
Li, Ziyuan
Han, Fangwei
Huang, Yajing
Qiu, Hengzhi
Zhu, Bing
He, Peiyang
contents We present Verified Multi-Agent Orchestration (VMAO), a framework that coordinates specialized LLM-based agents through a verification-driven iterative loop. Given a complex query, our system decomposes it into a directed acyclic graph (DAG) of sub-questions, executes them through domain-specific agents in parallel, verifies result completeness via LLM-based evaluation, and adaptively replans to address gaps. The key contributions are: (1) dependency-aware parallel execution over a DAG of sub-questions with automatic context propagation, (2) verification-driven adaptive replanning that uses an LLM-based verifier as an orchestration-level coordination signal, and (3) configurable stop conditions that balance answer quality against resource usage. On 25 expert-curated market research queries, VMAO improves answer completeness from 3.1 to 4.2 and source quality from 2.6 to 4.1 (1-5 scale) compared to a single-agent baseline, demonstrating that orchestration-level verification is an effective mechanism for multi-agent quality assurance.
format Preprint
id arxiv_https___arxiv_org_abs_2603_11445
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution
Zhang, Xing
Cui, Yanwei
Wang, Guanghui
Qiu, Wei
Li, Ziyuan
Han, Fangwei
Huang, Yajing
Qiu, Hengzhi
Zhu, Bing
He, Peiyang
Artificial Intelligence
Multiagent Systems
We present Verified Multi-Agent Orchestration (VMAO), a framework that coordinates specialized LLM-based agents through a verification-driven iterative loop. Given a complex query, our system decomposes it into a directed acyclic graph (DAG) of sub-questions, executes them through domain-specific agents in parallel, verifies result completeness via LLM-based evaluation, and adaptively replans to address gaps. The key contributions are: (1) dependency-aware parallel execution over a DAG of sub-questions with automatic context propagation, (2) verification-driven adaptive replanning that uses an LLM-based verifier as an orchestration-level coordination signal, and (3) configurable stop conditions that balance answer quality against resource usage. On 25 expert-curated market research queries, VMAO improves answer completeness from 3.1 to 4.2 and source quality from 2.6 to 4.1 (1-5 scale) compared to a single-agent baseline, demonstrating that orchestration-level verification is an effective mechanism for multi-agent quality assurance.
title Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution
topic Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2603.11445