2606.13598v1 Jun 11, 2026 cs.AI

Reward Modeling for Multi-Agent Orchestration

Zixuan Ke
Zixuan Ke
Citations: 345
h-index: 9
Semih Yavuz
Semih Yavuz
Citations: 2,800
h-index: 25
Shafiq Joty
Shafiq Joty
Citations: 816
h-index: 18
Vishal Venkataramani
Vishal Venkataramani
Citations: 5
h-index: 2
Haizhou Shi
Haizhou Shi
Citations: 524
h-index: 6
King Yeung Tsang
King Yeung Tsang
Citations: 0
h-index: 0
Hao Wang
Hao Wang
Citations: 404
h-index: 4
Zihao Zhao
Zihao Zhao
Citations: 6
h-index: 2

Multi-Agent Systems (MAS) built on Large Language Models (LLMs) require effective orchestration to coordinate specialized agents, yet training such orchestrators is hindered by limited supervision and high computational cost. We propose Orchestration Reward Modeling (OrchRM), a self-supervised framework for evaluating orchestration quality without human annotations. OrchRM leverages intermediate artifacts from multi-agent executions to construct win-lose pairs for Bradley-Terry reward model training. Unlike existing MAS test-time scaling and orchestrator training frameworks that rely on costly sub-agent rollouts, OrchRM operates directly at the orchestration level, enabling efficient and high-performing reward-guided orchestrator training and MAS test-time scaling. OrchRM improves training efficiency by up to 10x in token usage while improving MAS test-time scaling performance by up to 8% in accuracy. These gains consistently transfer across multiple domains, including mathematical reasoning, web-based question answering, and multi-hop reasoning, demonstrating orchestration-level reward modeling as a scalable direction for robust multi-agent orchestration. Code will be available at https://github.com/Wang-ML-Lab/OrchRM.

0 Citations
0 Influential
32.5 Altmetric
162.5 Score
Original PDF
0

No Analysis Report Yet

This paper hasn't been analyzed by Gemini yet.

Log in to request an AI analysis.

댓글

댓글을 작성하려면 로그인하세요.

아직 댓글이 없습니다. 첫 번째 댓글을 남겨보세요!