UNO Arena:用于评估大语言模型的序列决策能力
UNO Arena for Evaluating Sequential Decision-Making Capability of Large Language Models
AI总结:
该研究提出基于UNO卡牌游戏的UNO Arena评估框架,采用蒙特卡洛方法的新指标动态评测大语言模型的序列决策能力,并提出结合历史总结与策略反思的TUTRI玩家,显著提升了大语言模型的序列决策表现。
摘要:
序列决策指的是考虑环境动态性的算法,其中早期决策会影响后续决策。随着大语言模型(LLMs)在各类任务间展现出强大能力,我们不禁要问:当前的LLMs能否有效进行序列决策?为回答这一问题,我们基于卡牌游戏UNO提出了UNO Arena(UNO竞技场),用于评估LLMs的序列决策能力,并详细阐释了选择UNO的原因。在UNO Arena中,我们采用基于蒙特卡洛方法的新型指标,动态评估LLMs的序列决策能力。我们设置了随机玩家、基于DQN(深度Q网络)的强化学习玩家以及LLM玩家(例如GPT-4、Gemini-pro)开展对比测试。此外,为提升LLMs的序列决策能力,我们提出了TUTRI玩家,它能让LLMs结合游戏历史总结与游戏策略,对自身行动进行反思。大量实验表明,与原生LLM玩家相比,TUTRI玩家在序列决策性能上实现了显著突破。
英文摘要:
Sequential decision-making refers to algorithms that take into account the dynamics of the environment, where early decisions affect subsequent decisions. With large language models (LLMs) demonstrating powerful capabilities between tasks, we can't help but ask: Can Current LLMs Effectively Make Sequential Decisions? In order to answer this question, we propose the UNO Arena based on the card game UNO to evaluate the sequential decision-making capability of LLMs and explain in detail why we choose UNO. In UNO Arena, We evaluate the sequential decision-making capability of LLMs dynamically with novel metrics based Monte Carlo methods. We set up random players, DQN-based reinforcement learning players, and LLM players (e.g. GPT-4, Gemini-pro) for comparison testing. Furthermore, in order to improve the sequential decision-making capability of LLMs, we propose the TUTRI player, which can involves having LLMs reflect their own actions wtih the summary of game history and the game strategy. Numerous experiments demonstrate that the TUTRI player achieves a notable breakthrough in the performance of sequential decision-making compared to the vanilla LLM player.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。