Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Who is Undercover? Guiding LLMs to Explore Multi-Perspective Team Tactic in the Game

Ruiqi Dong, Zhixuan Liao, Guangwei Lai, Yuhan Ma, Danni Ma, Chenyou Fan

TL;DR

Preliminary results show that MPTT, combined with WIU, leverages LLMs' cognitive capabilities to create a decision-making framework that can simulate real society, and aids minority groups in communication and expression, promoting fairness and diversity in decision-making.

Abstract

Large Language Models (LLMs) are pivotal AI agents in complex tasks but still face challenges in open decision-making problems within complex scenarios. To address this, we use the language logic game ``Who is Undercover?'' (WIU) as an experimental platform to propose the Multi-Perspective Team Tactic (MPTT) framework. MPTT aims to cultivate LLMs' human-like language expression logic, multi-dimensional thinking, and self-perception in complex scenarios. By alternating speaking and voting sessions, integrating techniques like self-perspective, identity-determination, self-reflection, self-summary and multi-round find-teammates, LLM agents make rational decisions through strategic concealment and communication, fostering human-like trust. Preliminary results show that MPTT, combined with WIU, leverages LLMs' cognitive capabilities to create a decision-making framework that can simulate real society. This framework aids minority groups in communication and expression, promoting fairness and diversity in decision-making. Additionally, our Human-in-the-loop experiments demonstrate that LLMs can learn and align with human behaviors through interactive, indicating their potential for active participation in societal decision-making.

Who is Undercover? Guiding LLMs to Explore Multi-Perspective Team Tactic in the Game

TL;DR

Preliminary results show that MPTT, combined with WIU, leverages LLMs' cognitive capabilities to create a decision-making framework that can simulate real society, and aids minority groups in communication and expression, promoting fairness and diversity in decision-making.

Abstract

Large Language Models (LLMs) are pivotal AI agents in complex tasks but still face challenges in open decision-making problems within complex scenarios. To address this, we use the language logic game ``Who is Undercover?'' (WIU) as an experimental platform to propose the Multi-Perspective Team Tactic (MPTT) framework. MPTT aims to cultivate LLMs' human-like language expression logic, multi-dimensional thinking, and self-perception in complex scenarios. By alternating speaking and voting sessions, integrating techniques like self-perspective, identity-determination, self-reflection, self-summary and multi-round find-teammates, LLM agents make rational decisions through strategic concealment and communication, fostering human-like trust. Preliminary results show that MPTT, combined with WIU, leverages LLMs' cognitive capabilities to create a decision-making framework that can simulate real society. This framework aids minority groups in communication and expression, promoting fairness and diversity in decision-making. Additionally, our Human-in-the-loop experiments demonstrate that LLMs can learn and align with human behaviors through interactive, indicating their potential for active participation in societal decision-making.

Paper Structure

This paper contains 12 sections, 11 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. MPTT: A framework for reasoning game
  4. Game description and overall process
  5. Phase I: Reflections on the Speaking Session
  6. Phase II: Reflections on the Voting Session
  7. Experiment
  8. Baselines and our approach
  9. Evaluation of game metrics
  10. LLM and human collaborative reasoning
  11. Advanced tactics and generalization ability
  12. Conclusion

Figures (4)

  • Figure 1: Presentation of “Who is Undercover? " (WIU) in the framework of Multi-Perspective Team Tactic (MPTT). The game alternates between reflection and action parts in the speaking and voting sessions, contains a series of techniques, and at the end of each session, a summary is generated to serve as a reference for the players to assists them in making decisions.
  • Figure 2: Improvement with MPTT. MPTT is significantly improved in three areas.
  • Figure 3: Ablation for statistics. The undercover team is significantly enhanced under MPTT over Baseline, effectively cutting down civilian forces. Ablation studies in (b) and (c) demonstrate the effectiveness of multidimensional self-reflection, global historical speech and self-summary.
  • Figure 4: Advanced tactics. Two examples emerge in WIU.