TY - GEN
T1 - CraftUtopia
T2 - 25th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2026
AU - Fu, Wanli
AU - Li, Hao
AU - Ren, Siyue
AU - Xing, Chenxi
AU - Chen, Yang
AU - Chu, Chen
AU - Wang, Zhen
AU - Hu, Shuyue
N1 - Publisher Copyright:
© 2026 International Foundation for Autonomous Agents and Multiagent Systems.
PY - 2026/5/24
Y1 - 2026/5/24
N2 - LLM-based multi-agent construction is a growing research area but prior works still suffer from low success rates for construction, reliance on predefined templates, and poor scalability as the number of agents increases. To address these challenges, we present CraftUtopia, an LLM-based multi-agent system (MAS) that constructs 3D structures in Minecraft from a single 2D reference image. CraftUtopia operates in two stages: Design, which converts the image into a Minecraft-compatible 3D blueprint, and Build, which decomposes the blueprint into spatially disjoint subtasks for parallel execution. CraftUtopia scales efficiently via two mechanisms: (i) hierarchical coordination, which organizes agents in a manager-foreman-worker hierarchy to separate responsibilities, and (ii) skill acquisition, which distills recurring action sequences into a shared skill library to reduce repeated LLM replanning. Across three representative builds, CraftUtopia achieves 100% success over five trials using only 2D inputs, scales effectively with more workers, and exhibits emergent human-like behaviors. A full demo is available at: https://github.com/craftutopia-demo/CraftUtopia.
AB - LLM-based multi-agent construction is a growing research area but prior works still suffer from low success rates for construction, reliance on predefined templates, and poor scalability as the number of agents increases. To address these challenges, we present CraftUtopia, an LLM-based multi-agent system (MAS) that constructs 3D structures in Minecraft from a single 2D reference image. CraftUtopia operates in two stages: Design, which converts the image into a Minecraft-compatible 3D blueprint, and Build, which decomposes the blueprint into spatially disjoint subtasks for parallel execution. CraftUtopia scales efficiently via two mechanisms: (i) hierarchical coordination, which organizes agents in a manager-foreman-worker hierarchy to separate responsibilities, and (ii) skill acquisition, which distills recurring action sequences into a shared skill library to reduce repeated LLM replanning. Across three representative builds, CraftUtopia achieves 100% success over five trials using only 2D inputs, scales effectively with more workers, and exhibits emergent human-like behaviors. A full demo is available at: https://github.com/craftutopia-demo/CraftUtopia.
KW - Cooperation
KW - Large Language Model
KW - Multi-agent Construction
UR - https://www.scopus.com/pages/publications/105041455828
U2 - 10.65109/PUIJ7087
DO - 10.65109/PUIJ7087
M3 - 会议稿件
AN - SCOPUS:105041455828
T3 - AAMAS 2026 - Proceedings of the 25th International Conference on Autonomous Agents and Multiagent Systems
SP - 4140
EP - 4142
BT - AAMAS 2026 - Proceedings of the 25th International Conference on Autonomous Agents and Multiagent Systems
PB - Association for Computing Machinery, Inc
Y2 - 25 May 2026 through 29 May 2026
ER -