OneFlowSBI: One Model, Many Queries for Simulation-Based Inference
This work addresses the problem of flexible and efficient inference for researchers and practitioners in fields like computational science, offering a general-purpose solution that is incremental in improving upon existing specialized methods.
The paper tackles the challenge of simulation-based inference by introducing OneFlowSBI, a unified framework that learns a single generative model to handle multiple inference tasks without retraining, achieving competitive performance on benchmarks and real-world problems with efficient sampling and robustness to noisy data.
We introduce \textit{OneFlowSBI}, a unified framework for simulation-based inference that learns a single flow-matching generative model over the joint distribution of parameters and observations. Leveraging a query-aware masking distribution during training, the same model supports multiple inference tasks, including posterior sampling, likelihood estimation, and arbitrary conditional distributions, without task-specific retraining. We evaluate \textit{OneFlowSBI} on ten benchmark inference problems and two high-dimensional real-world inverse problems across multiple simulation budgets. \textit{OneFlowSBI} is shown to deliver competitive performance against state-of-the-art generalized inference solvers and specialized posterior estimators, while enabling efficient sampling with few ODE integration steps and remaining robust under noisy and partially observed data.