Agentic Design of Compositional Machines
By: Wenqian Zhang, Weiyang Liu, Zhen Liu
Potential Business Impact:
AI learns to build machines from parts.
The design of complex machines stands as both a marker of human intelligence and a foundation of engineering practice. Given recent advances in large language models (LLMs), we ask whether they, too, can learn to create. We approach this question through the lens of compositional machine design: a task in which machines are assembled from standardized components to meet functional demands like locomotion or manipulation in a simulated physical environment. To support this investigation, we introduce BesiegeField, a testbed built on the machine-building game Besiege, which enables part-based construction, physical simulation and reward-driven evaluation. Using BesiegeField, we benchmark state-of-the-art LLMs with agentic workflows and identify key capabilities required for success, including spatial reasoning, strategic assembly, and instruction-following. As current open-source models fall short, we explore reinforcement learning (RL) as a path to improvement: we curate a cold-start dataset, conduct RL finetuning experiments, and highlight open challenges at the intersection of language, machine design, and physical reasoning.
Similar Papers
Agentic Design of Compositional Machines
Artificial Intelligence
Computers learn to build machines from parts.
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Artificial Intelligence
Lets AI learn to make smart choices.
AgentCoMa: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World Scenarios
Computation and Language
AI struggles combining thinking and math skills.