Model Capacity Frontier: A Development Framework for LLM-Supported Social Simulation

DOI10.48550/ARXIV.2610.24168MaRDI QIDQ7400447arXivFDO

Authors Hirokazu Shirado, Patrick Zhao, Yuxuan Li

Publication date 16 October 2026

Published in arXiv

Tldr We present Model Capacity Frontier (MCF), a framework that characterizes the empirical performance attainable by a specified architecture across its admissible configurations. This allows for the identification of which dimensions of model capacity are improved or limited when integrating LLMs into ABMs.

Full work available at URL https://doi.org/10.48550/ARXIV.2610.24168

arXiv classification

cs.AIcs.CYcs.MA

Cites work

• A fast marching level set method for monotonically advancing fronts.
• Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
• Bayesian calibration of computer models. (With discussion)
• Mohcine Chraibi
• On the distribution of points in a cube and the approximate evaluation of integrals
• Pattern-Oriented Modeling of Agent-Based Complex Systems: Lessons from Ecology
• Saving human lives: what complexity science and information systems can contribute
• The fundamental diagram of pedestrian movement revisited

This page was built for publication: Model Capacity Frontier: A Development Framework for LLM-Supported Social Simulation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7400447)