Influencing LLM Multi-Agent Dialogue via Policy-Parameterized Prompts

2026-03-10

AI Summary

This work addresses the lack of systematic behavioral control mechanisms in existing large language model–based multi-agent dialogue systems, which often rely on ad hoc prompt engineering. The authors propose a lightweight policy framework that, for the first time, treats prompts as parameterizable policy actions. By introducing a state-aware dynamic prompting mechanism, the framework enables controllable intervention in multi-agent conversations without requiring model fine-tuning. It comprises five components that jointly generate guiding prompts in real time based on the current dialogue state of each agent. Experimental results on two public-issue discussion scenarios demonstrate significant improvements across key metrics, including responsiveness, rebuttal quality, evidence utilization, non-redundancy, and stance evolution.

Technology Category

• Natural Language Processing: Prompt Engineering / Prompting
• Multiagent Systems: Adversarial Agents
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Large Language Models (LLMs) have emerged as a new paradigm for multi-agent systems. However, existing research on the behaviour of LLM-based multi-agents relies on ad hoc prompts and lacks a principled policy perspective. Different from reinforcement learning, we investigate whether prompt-as-action can be parameterized so as to construct a lightweight policy which consists of a sequence of state-action pairs to influence conversational behaviours without training. Our framework regards prompts as actions executed by LLMs, and dynamically constructs prompts through five components based on the current state of the agent. To test the effectiveness of parameterized control, we evaluated the dialogue flow based on five indicators: responsiveness, rebuttal, evidence usage, non-repetition, and stance shift. We conduct experiments using different LLM-driven agents in two discussion scenarios related to the general public and show that prompt parameterization can influence the dialogue dynamics. This result shows that policy-parameterised prompts offer a simple and effective mechanism to influence the dialogue process, which will help the research of multi-agent systems in the direction of social simulation.

Problem

• LLM multi-agent systems
• prompt design
• dialogue dynamics
• policy parameterization
• conversational behavior

Innovation

• policy-parameterized prompts
• multi-agent dialogue
• prompt-as-action
• LLM-based agents
• dialogue dynamics