HomeResearch LibraryVoyager: An Open-Ended Embodied Agent with Large Langua…
system architectureChapter 2arXiv · 2023

Voyager: An Open-Ended Embodied Agent with Large Language Models

Guanzhi Wang (NVIDIA), Yuqi Xie (Caltech), Yunfan Jiang (Stanford)

Abstract

We present Voyager, the first LLM-powered embodied lifelong learning agent in Minecraft that continuously explores the world, acquires diverse skills, and makes novel discoveries without human intervention.

Key Contributions

  • →Lifelong learning agent architecture
  • →Skill library for reuse
  • →Automatic curriculum generation
Eigenvector Commentary

Voyager's skill library concept is directly applicable to enterprise agents. The idea of building a reusable library of verified, tested skills — rather than regenerating procedures from scratch each time — is a key architectural pattern for Zone III efficiency.

Topics

lifelong learningskill acquisitionembodied agentslong-horizon planning