Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions
Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions
复制标题
用于在线决策的 Auto-GPT：基准和附加意见
DOI：
10.48550/arxiv.2306.02224
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Yunzhong He
中科院分区：
文献类型：
--
作者：
Hui Yang;Sifu Yue;Yunzhong He
Auto-GPT是一种自主代理，它利用了最近在适应大型语言模型(LLM)用于决策任务方面的进展。虽然人们对Auto-GPT定型代理的兴趣与日俱增，但关于Auto-GPT在解决现实世界决策任务中的有效性和灵活性的问题仍然存在。它在现实世界中参与的能力有限，而且没有基准，这些都是造成这些不确定性的原因。在本文中，我们提出了一个全面的基准研究，自动GPT风格的代理在模拟现实世界场景的决策任务。我们的目标是对这个问题有更深入的了解，并了解基于GPT的代理的适应性。我们比较了流行的LLMS，如GPT-4，GPT-3.5，克劳德和维古纳在Auto-GPT风格的决策任务中的表现。此外，我们还引入了附加意见算法，这是一种简单而有效的方法，它将基于监督/模仿的学习者结合到Auto-GPT方案中。这种方法实现了轻量级监督学习，而不需要对基本LLM进行微调。我们通过仔细的基线比较和消融研究证明，附加意见算法显著提高了在线决策基准测试的性能，包括Webshop和ALFWorld。
Auto-GPT is an autonomous agent that leverages recent advancements in adapting Large Language Models (LLMs) for decision-making tasks. While there has been a growing interest in Auto-GPT stypled agents, questions remain regarding the effectiveness and flexibility of Auto-GPT in solving real-world decision-making tasks. Its limited capability for real-world engagement and the absence of benchmarks contribute to these uncertainties. In this paper, we present a comprehensive benchmark study of Auto-GPT styled agents in decision-making tasks that simulate real-world scenarios. Our aim is to gain deeper insights into this problem and understand the adaptability of GPT-based agents. We compare the performance of popular LLMs such as GPT-4, GPT-3.5, Claude, and Vicuna in Auto-GPT styled decision-making tasks. Furthermore, we introduce the Additional Opinions algorithm, an easy and effective method that incorporates supervised/imitation-based learners into the Auto-GPT scheme. This approach enables lightweight supervised learning without requiring fine-tuning of the foundational LLMs. We demonstrate through careful baseline comparisons and ablation studies that the Additional Opinions algorithm significantly enhances performance in online decision-making benchmarks, including WebShop and ALFWorld.
DOI：
--
发表时间：
2022-01
期刊：
ArXiv
影响因子：
--
作者：
Jason Wei;Xuezhi Wang;Dale Schuurmans;Maarten Bosma;E. Chi;F. Xia;Quoc Le;Denny Zhou
通讯作者：
Jason Wei;Xuezhi Wang;Dale Schuurmans;Maarten Bosma;E. Chi;F. Xia;Quoc Le;Denny Zhou