How Far Are We on the Decision-Making of LLMs? Evaluating LLMs' Gaming Ability in Multi-Agent Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Jen-tse, Li, Eric John, Lam, Man Ho, Liang, Tian, Wang, Wenxuan, Yuan, Youliang, Jiao, Wenxiang, Wang, Xing, Tu, Zhaopeng, Lyu, Michael R.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915183478177792
author Huang, Jen-tse
Li, Eric John
Lam, Man Ho
Liang, Tian
Wang, Wenxuan
Yuan, Youliang
Jiao, Wenxiang
Wang, Xing
Tu, Zhaopeng
Lyu, Michael R.
author_facet Huang, Jen-tse
Li, Eric John
Lam, Man Ho
Liang, Tian
Wang, Wenxuan
Yuan, Youliang
Jiao, Wenxiang
Wang, Xing
Tu, Zhaopeng
Lyu, Michael R.
contents Decision-making is a complex process requiring diverse abilities, making it an excellent framework for evaluating Large Language Models (LLMs). Researchers have examined LLMs' decision-making through the lens of Game Theory. However, existing evaluation mainly focus on two-player scenarios where an LLM competes against another. Additionally, previous benchmarks suffer from test set leakage due to their static design. We introduce GAMA($γ$)-Bench, a new framework for evaluating LLMs' Gaming Ability in Multi-Agent environments. It includes eight classical game theory scenarios and a dynamic scoring scheme specially designed to quantitatively assess LLMs' performance. $γ$-Bench allows flexible game settings and adapts the scoring system to different game parameters, enabling comprehensive evaluation of robustness, generalizability, and strategies for improvement. Our results indicate that GPT-3.5 demonstrates strong robustness but limited generalizability, which can be enhanced using methods like Chain-of-Thought. We also evaluate 13 LLMs from 6 model families, including GPT-3.5, GPT-4, Gemini, LLaMA-3.1, Mixtral, and Qwen-2. Gemini-1.5-Pro outperforms others, scoring of $69.8$ out of $100$, followed by LLaMA-3.1-70B ($65.9$) and Mixtral-8x22B ($62.4$). Our code and experimental results are publicly available at https://github.com/CUHK-ARISE/GAMABench.
format Preprint
id arxiv_https___arxiv_org_abs_2403_11807
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle How Far Are We on the Decision-Making of LLMs? Evaluating LLMs' Gaming Ability in Multi-Agent Environments
Huang, Jen-tse
Li, Eric John
Lam, Man Ho
Liang, Tian
Wang, Wenxuan
Yuan, Youliang
Jiao, Wenxiang
Wang, Xing
Tu, Zhaopeng
Lyu, Michael R.
Artificial Intelligence
Computation and Language
Decision-making is a complex process requiring diverse abilities, making it an excellent framework for evaluating Large Language Models (LLMs). Researchers have examined LLMs' decision-making through the lens of Game Theory. However, existing evaluation mainly focus on two-player scenarios where an LLM competes against another. Additionally, previous benchmarks suffer from test set leakage due to their static design. We introduce GAMA($γ$)-Bench, a new framework for evaluating LLMs' Gaming Ability in Multi-Agent environments. It includes eight classical game theory scenarios and a dynamic scoring scheme specially designed to quantitatively assess LLMs' performance. $γ$-Bench allows flexible game settings and adapts the scoring system to different game parameters, enabling comprehensive evaluation of robustness, generalizability, and strategies for improvement. Our results indicate that GPT-3.5 demonstrates strong robustness but limited generalizability, which can be enhanced using methods like Chain-of-Thought. We also evaluate 13 LLMs from 6 model families, including GPT-3.5, GPT-4, Gemini, LLaMA-3.1, Mixtral, and Qwen-2. Gemini-1.5-Pro outperforms others, scoring of $69.8$ out of $100$, followed by LLaMA-3.1-70B ($65.9$) and Mixtral-8x22B ($62.4$). Our code and experimental results are publicly available at https://github.com/CUHK-ARISE/GAMABench.
title How Far Are We on the Decision-Making of LLMs? Evaluating LLMs' Gaming Ability in Multi-Agent Environments
topic Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2403.11807