SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.16291 · 2023

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

Anima Anandkumar, U Austin, Yuke Zhu, Caltech, Guanzhi Wang, Chaowei Xiao, Yunfan Jiang, Ajay Mandlekar, Linxi, Jim Fan, Yuqi Xie, U Madison

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 6 functions out of this paper's own repositories and ran 0 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
MineDojo/Voyager canonical 0 of 6
FunctionStatusWhere it lives
extract_char_position Not yet run MineDojo/Voyager/voyager/utils/json_utils.py
code served (permissive licence) · get_code("05d3e32da2834d25")
is_sequence Not yet run MineDojo/Voyager/voyager/utils/file_utils.py
code served (permissive licence) · get_code("377feaa4a69f18d2")
json_dumps Not yet run MineDojo/Voyager/voyager/utils/json_utils.py
code served (permissive licence) · get_code("2db34269a980a57d")
json_loads Not yet run MineDojo/Voyager/voyager/utils/json_utils.py
code served (permissive licence) · get_code("133376979a8abde5")
pack_varargs Not yet run MineDojo/Voyager/voyager/utils/file_utils.py
code served (permissive licence) · get_code("4926aacb60add226")
utf_open Not yet run MineDojo/Voyager/voyager/utils/file_utils.py
code served (permissive licence) · get_code("edd852217838fe9c")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We introduce 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. VOYAGER consists of three key components: 1) an automatic curriculum that maximizes exploration, 2) an ever-growing skill library of executable code for storing and retrieving complex behaviors, and 3) a new iterative prompting mechanism that incorporates environment feedback, execution errors, and self-verification for program improvement. VOYAGER interacts with GPT-4 via blackbox queries, which bypasses the need for model parameter fine-tuning. The skills developed by VOYAGER are temporally extended, interpretable, and compositional, which compounds the agent's abilities rapidly and alleviates catastrophic forgetting. Empirically, VOYAGER shows strong in-context lifelong learning capability and exhibits exceptional proficiency in playing Minecraft. It obtains 3.3× more unique items, travels 2.3× longer distances, and unlocks key tech tree milestones up to 15.3× faster than prior SOTA. VOYAGER is able to utilize the learned skill library in a new Minecraft world to solve novel tasks from scratch, while other techniques struggle to generalize.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2305.16291")
get_code_for_paper("2305.16291")
have("2305.16291")

Connect an agent — have() is free.