{"ID":2892472,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.16069","arxiv_id":"2507.16069","title":"Interpreting CFD Surrogates through Sparse Autoencoders","abstract":"Learning-based surrogate models have become a practical alternative to high-fidelity CFD solvers, but their latent representations remain opaque and hinder adoption in safety-critical or regulation-bound settings. This work introduces a posthoc interpretability framework for graph-based surrogate models used in computational fluid dynamics (CFD) by leveraging sparse autoencoders (SAEs). By obtaining an overcomplete basis in the node embedding space of a pretrained surrogate, the method extracts a dictionary of interpretable latent features. The approach enables the identification of monosemantic concepts aligned with physical phenomena such as vorticity or flow structures, offering a model-agnostic pathway to enhance explainability and trustworthiness in CFD applications.","short_abstract":"Learning-based surrogate models have become a practical alternative to high-fidelity CFD solvers, but their latent representations remain opaque and hinder adoption in safety-critical or regulation-bound settings. This work introduces a posthoc interpretability framework for graph-based surrogate models used in computa...","url_abs":"https://arxiv.org/abs/2507.16069","url_pdf":"https://arxiv.org/pdf/2507.16069v1","authors":"[\"Yeping Hu\",\"Shusen Liu\"]","published":"2025-07-21T21:09:45Z","proceeding":"cs.CE","tasks":"[\"cs.CE\",\"cs.LG\"]","methods":"[]","has_code":false}