AgentGroupChat: An Interactive Group Chat Simulacra For Better Eliciting Emergent Behavior

TL;DR

AgentGroupChat is introduced, a simulation that delves into the complex role of language in shaping collective behavior through interactive debate scenarios, and introduces the Verbal Strategist Agent, utilizing large language models to enhance interaction strategies by incorporating elements of persona and action.

Abstract

Language significantly influences the formation and evolution of Human emergent behavior, which is crucial in understanding collective intelligence within human societies. Considering that the study of how language affects human behavior needs to put it into the dynamic scenarios in which it is used, we introduce AgentGroupChat in this paper, a simulation that delves into the complex role of language in shaping collective behavior through interactive debate scenarios. Central to this simulation are characters engaging in dynamic conversation interactions. To enable simulation, we introduce the Verbal Strategist Agent, utilizing large language models to enhance interaction strategies by incorporating elements of persona and action. We set four narrative scenarios based on AgentGroupChat to demonstrate the simulation's capacity to mimic complex language use in group dynamics. Evaluations focus on aligning agent behaviors with human expectations and the emergence of collective behaviors within the simulation. Results reveal that emergent behaviors materialize from a confluence of factors: a conducive environment for extensive information exchange, characters with diverse traits, high linguistic comprehension, and strategic adaptability. During discussions on ``the impact of AI on humanity'' in AgentGroupChat simulation, philosophers commonly agreed that ``AI could enhance societal welfare with judicious limitations'' and even come to a conclusion that ``the essence of true intelligence encompasses understanding the necessity to constrain self abilities''. Additionally, in the competitive domain of casting for primary roles in films in AgentGroupChat, certain actors were ready to reduce their remuneration or accept lesser roles, motivated by their deep-seated desire to contribute to the project.

Figures (8)

• Figure 1: The four components of the AgentGroupChat environment, including Character (green part), Resource (red part), Progress (gray part), and Information (blue part). The main components of the Verb Strategist Agent are Persona and Action. Persona consists of Scratch, Relationship Judgement, Belief, and Memory. Action consists of Planning, Choosing, Speaking, Summarization, Reflection, and Voting. Progress consists of the Update Stage, Private Chatting Stage, Confidential Meeting Stage, Group Chatting Stage, and Settle Stage. Information consists of Progress Description, Resources and Characters Description and Action History.
• Figure 2: The memory structure and memory flow between each action of a VS Agent. Logan Roy is an example.
• Figure 3: The proposed structural causal model (SCM) to decompose the variants that affect the final emergent behavior.
• Figure 4: The case of how VS Agent behaves Shiv Roy without each key component in its architecture.
• Figure 5: An illustration of strategic communication and power dynamics among key participants in Inheritance Disputes. Private chatting, confidential meetings, and group discussions highlight a complex network of alliances and conflicts, driving the development of control and influence within the Roy family.