arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.08367cs.AIcs.CLcs.CVcs.LG
收藏

MCU:一个面向开放结局游戏智能体的评估框架

MCU: An Evaluation Framework for Open-Ended Game Agents

Institute for Artificial Intelligence, Peking University(北京大学人工智能研究院);Tencent AI Lab(腾讯人工智能实验室)

Xinyue Zheng, Haowei Lin, Kaichen He, Zihao Wang, Zilong Zheng, Yitao Liang

AI总结:

本文提出 Minecraft Universe(MCU)评估框架,通过 3,452 个可组合原子任务、无限任务生成机制和与人类评分 91.5% 一致的评估方法,揭示现有基础智能体在开放结局任务中的不足。

摘要:

开发能够与开放世界环境交互以解决多样化任务的 AI 智能体是一个极具吸引力的挑战。然而,评估此类开放结局智能体仍然困难,现有基准面临可扩展性限制。为解决这一问题,我们提出了 Minecraft Universe(MCU),一个构建在开放世界视频游戏 Minecraft 中的综合评估框架。MCU 包含三个关键组成部分:(1)一个不断扩展的、由 3,452 个可组合原子任务构成的集合,涵盖 11 个主要类别和 41 个子类别的挑战;(2)一个任务组合机制,能够生成无限多样且难度各异的任务;(3)一个通用评估框架,在开放结局任务评估中实现了与人类评分 91.5% 的一致性。实验结果表明,即使是最先进的基础智能体也难以应对任务日益增长的多样性和复杂性。这些发现凸显了 MCU 作为一个稳健基准的必要性,以推动开放结局环境中 AI 智能体开发的进步。我们的评估代码和脚本可在 https://github.com/CraftJarvis/MCU 获取。

英文摘要:

Developing AI agents capable of interacting with open-world environments to solve diverse tasks is a compelling challenge. However, evaluating such open-ended agents remains difficult, with current benchmarks facing scalability limitations. To address this, we introduce Minecraft Universe (MCU), a comprehensive evaluation framework set within the open-world video game Minecraft. MCU incorporates three key components: (1) an expanding collection of 3,452 composable atomic tasks that encompasses 11 major categories and 41 subcategories of challenges; (2) a task composition mechanism capable of generating infinite diverse tasks with varying difficulty; and (3) a general evaluation framework that achieves 91.5\% alignment with human ratings for open-ended task assessment. Empirical results reveal that even state-of-the-art foundation agents struggle with the increasing diversity and complexity of tasks. These findings highlight the necessity of MCU as a robust benchmark to drive progress in AI agent development within open-ended environments. Our evaluation code and scripts are available at https://github.com/CraftJarvis/MCU.

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。

↑