[2603.20520] CogFormer: Learn All Your Models Once

Statistics > Machine Learning arXiv:2603.20520 (stat) [Submitted on 20 Mar 2026] Title:CogFormer: Learn All Your Models Once Authors:Jerry M. Huang, Lukas Schumacher, Niek Stevenson, Stefan T. Radev View a PDF of the paper titled CogFormer: Learn All Your Models Once, by Jerry M. Huang and Lukas Schumacher and Niek Stevenson and Stefan T. Radev View PDF Abstract:Simulation-based inference (SBI) with neural networks has accelerated and transformed cognitive modeling workflows. SBI enables modelers to fit complex models that were previously difficult or impossible to estimate, while also allowing rapid estimation across large numbers of datasets. However, the utility of SBI for iterating over varying modeling assumptions remains limited: changing parameterizations, generative functions, priors, and design variables all necessitate model retraining and hence diminish the benefits of amortization. To address these issues, we pilot a meta-amortized framework for cognitive modeling which we nickname the CogFormer. Our framework trains a transformer-based architecture that remains valid across a combinatorial number of structurally similar models, allowing for changing data types, parameters, design matrices, and sample sizes. We present promising quantitative results across families of decision-making models for binary, multi-alternative, and continuous responses. Our evaluation suggests that CogFormer can accurately estimate parameters across model families with a minimal amort...