Control-Data Flow Separation: Stable Prompt Optimization in Multi-Agent LLMs

2026-08-31

AI Summary

论文提出控制-数据流分离方法，解决多代理LLM系统中提示优化导致协议失效问题，通过将执行关键控制与任务相关语言分离，提高系统性能。

Abstract

Prompt optimization can improve multi-agent LLM systems, but the prompts being optimized often serve two entangled roles: generating task-relevant content and specifying execution-critical protocols, such as message routing, output formatting, and termination signals, on which the underlying code relies. As a result, a prompt edit intended to improve content generation can inadvertently corrupt the protocol and cause the entire agent pipeline to fail. Our key observation is that these two roles have different representations: execution protocols are typically structured, while task-relevant content is usually expressed in unstructured language. Based on this, we propose control-data flow separation, where execution-critical control is represented as typed, validated program objects, while task-relevant language remains the optimizable data flow for agent communication. This design allows optimizers to improve multi-agent behavior without exposing the routing or formatting interface to prompt drift. Across synthetic reasoning, collaborative review generation, and insurance rating workflows, our framework empirically achieves 100% eventual protocol validity while consistently improving task performance.

Problem

• Prompt Optimization
• Multi-Agent LLMs
• Execution Protocols

Innovation

• control-data flow separation
• prompt optimization
• multi-agent LLMs
• protocol validity