文章目录 1 Introduction2 Related Works2.1 Human-AI Interaction2.2 Belivable Proxies for Human Behavior2.3 Large Language Model and Human Behavior 3 Generative agent behavior and interaction(行为与交互)3.1 Agent Avatar and Communicatio…
UFO:A UI-Focused Agent for Windows OS Interaction 前言AbstractMotivationMethodsExperimentConclusion 前言
Windows客户端第一个JARVIS,利用GPT4 Vision识别截图信息辅助智能体自动化执行操作,作为微软大肆宣传的一篇工作,其…
本文是LLM系列文章,针对《Unleashing the Power of Graph Learning through LLM-based Autonomous Agents》的翻译。 通过基于LLM的自动Agent释放图学习的力量 摘要1 引言2 相关工作3 方法4 实验5 结论 摘要
图结构数据在现实世界中广泛存在和应用,而以…
OS-Copilot: Towards Generalist Computer Agents with Self-Improvement 前言ABSTRACT1 INTRODUCTION2 THE OS-COPILOT FRAMEWORK2.1 PLANNER2.2 CONFIGURATOR2.2.1 DECLARATIVE MEMORY2.2.2 PROCEDURAL MEMORY2.2.3 WORKING MEMORY 2.3 ACTOR 3 THE FRIDAY AGENT3.1 A RUNNIN…
论文标题:MemGPT: Towards LLMs as Operating Systems 论文作者:Charles Packer, Vivian Fang, Shishir G. Patil, Kevin Lin, Sarah Wooders, Joseph E. Gonzalez (UC Berkeley) 论文原文:https://arxiv.org/abs/2310.08560 论文出处&#x…
REACT: SYNERGIZING REASONING AND ACTING IN LANGUAGE MODELS 前言ABSTRACT1 INTRODUCTION2 REACT: SYNERGIZING REASONING ACTING3 KNOWLEDGE-INTENSIVE REASONING TASKS3.1 SETUP3.2 METHODS3.3 RESULTS AND OBSERVATIONS 4 DECISION MAKING TASKS5 RELATED WORK6 CONCLUSI…