G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibration

Samuel Holt, Max Ruiz Luyten, Antonin Berthon, Mihaela Van Der Schaar
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:23513-23561, 2025.

Abstract

Constructing robust simulators is essential for asking "what if?" questions and guiding policy in critical domains like healthcare and logistics. However, existing methods often struggle, either failing to generalize beyond historical data or, when using Large Language Models (LLMs), suffering from inaccuracies and poor empirical alignment. We introduce G-Sim, a hybrid framework that automates simulator construction by synergizing LLM-driven structural design with rigorous empirical calibration. G-Sim employs an LLM in an iterative loop to propose and refine a simulator’s core components and causal relationships, guided by domain knowledge. This structure is then grounded in reality by estimating its parameters using flexible calibration techniques. Specifically, G-Sim can leverage methods that are both likelihood-free and gradient-free with respect to the simulator, such as gradient-free optimization for direct parameter estimation or simulation-based inference for obtaining a posterior distribution over parameters. This allows it to handle non-differentiable and stochastic simulators. By integrating domain priors with empirical evidence, G-Sim produces reliable, causally-informed simulators, mitigating data-inefficiency and enabling robust system-level interventions for complex decision-making.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-holt25b, title = {G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibration}, author = {Holt, Samuel and Ruiz Luyten, Max and Berthon, Antonin and Van Der Schaar, Mihaela}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {23513--23561}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/holt25b/holt25b.pdf}, url = {https://proceedings.mlr.press/v267/holt25b.html}, abstract = {Constructing robust simulators is essential for asking "what if?" questions and guiding policy in critical domains like healthcare and logistics. However, existing methods often struggle, either failing to generalize beyond historical data or, when using Large Language Models (LLMs), suffering from inaccuracies and poor empirical alignment. We introduce G-Sim, a hybrid framework that automates simulator construction by synergizing LLM-driven structural design with rigorous empirical calibration. G-Sim employs an LLM in an iterative loop to propose and refine a simulator’s core components and causal relationships, guided by domain knowledge. This structure is then grounded in reality by estimating its parameters using flexible calibration techniques. Specifically, G-Sim can leverage methods that are both likelihood-free and gradient-free with respect to the simulator, such as gradient-free optimization for direct parameter estimation or simulation-based inference for obtaining a posterior distribution over parameters. This allows it to handle non-differentiable and stochastic simulators. By integrating domain priors with empirical evidence, G-Sim produces reliable, causally-informed simulators, mitigating data-inefficiency and enabling robust system-level interventions for complex decision-making.} }
Endnote
%0 Conference Paper %T G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibration %A Samuel Holt %A Max Ruiz Luyten %A Antonin Berthon %A Mihaela Van Der Schaar %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-holt25b %I PMLR %P 23513--23561 %U https://proceedings.mlr.press/v267/holt25b.html %V 267 %X Constructing robust simulators is essential for asking "what if?" questions and guiding policy in critical domains like healthcare and logistics. However, existing methods often struggle, either failing to generalize beyond historical data or, when using Large Language Models (LLMs), suffering from inaccuracies and poor empirical alignment. We introduce G-Sim, a hybrid framework that automates simulator construction by synergizing LLM-driven structural design with rigorous empirical calibration. G-Sim employs an LLM in an iterative loop to propose and refine a simulator’s core components and causal relationships, guided by domain knowledge. This structure is then grounded in reality by estimating its parameters using flexible calibration techniques. Specifically, G-Sim can leverage methods that are both likelihood-free and gradient-free with respect to the simulator, such as gradient-free optimization for direct parameter estimation or simulation-based inference for obtaining a posterior distribution over parameters. This allows it to handle non-differentiable and stochastic simulators. By integrating domain priors with empirical evidence, G-Sim produces reliable, causally-informed simulators, mitigating data-inefficiency and enabling robust system-level interventions for complex decision-making.
APA
Holt, S., Ruiz Luyten, M., Berthon, A. & Van Der Schaar, M.. (2025). G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibration. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:23513-23561 Available from https://proceedings.mlr.press/v267/holt25b.html.

Related Material