Return
Can large language models implement agent-based models? An ODD-based replication study
DOI:10.1016/j.ecolmodel.2026.111624.png)
Abstract
En 中文
• LLMs can generate ABMs from ODD, but reliable replication remains uneven. • Statistical validation is essential, as executable code may still be incorrect. • GPT-4.1 most consistently reproduced PPHPC dynamics with good efficiency. • Successful implementations differed substantially in runtime and maintainability. • Results clarify current opportunities and limits for reproducible ABM workflows.
Keywords:
Large Language Models
Agent-Based Models
Replication Study
ODD
Statistical Validation
Journal
IF:
3.2
Papers:
626
Citations:
2.1W

