Influencing LLM Multi-Agent Dialogue via Policy-Parameterized Prompts
On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 17 of its 31 references that are in the map. First posted on arXiv 10 Mar 2026.
Builds on
17 of 31 references.
Closest ancestors · 1
Same direction, newest first
Other references · 16
- 2024The Llama 3 Herd of Modelscited by 14,715
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
- 2025Qwen3 Technical Reportcited by 8,820
- 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
- 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
- 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,463
- 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 1,011
- 2023Reasoning with Language Model is Planning with World Modelcited by 861
- 2022MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledgecited by 478
- 2022Social Simulacra: Creating Populated Prototypes for Social Computing Systemscited by 301
- 2023S3: Social-network Simulation System with Large Language Model-Empowered Agentscited by 233
- 2025AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Societycited by 186
4 more
- 2023Language Agents with Reinforcement Learning for Strategic Play in the Werewolf Gamecited by 152
- 2024Emergent social conventions and collective bias in LLM populationscited by 123
- 2022A Survey of Multi-Agent Deep Reinforcement Learning with Communicationcited by 104
- 2023AgentSims: An Open-Source Sandbox for Large Language Model Evaluationcited by 88
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Prompt reinforcing for long-term planning of large language models
- 2023Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents
- 2024Exploring and Controlling Diversity in LLM-Agent Conversation
- 2026Should LLM Agents Decide in Social Simulations? Comparing Finite-State and LLM-Based Decision Policies
- 2024Reinforcement Learning Problem Solving with Large Language Models
- 2025Evaluating LLM-based Agents for Multi-Turn Conversations: A Survey
- 2023Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs
- 2023Prompting and Evaluating Large Language Models for Proactive Dialogues: Clarification, Target-guided, and Non-collaboration