Nouveau Recherche PDF, HTML, DOCX et bien d'autres formats.

Prépublication · 2026

Evaluating LLMs for Translating Legacy Fortran Atmospheric Physics to GPU-Accelerated JAX

Iris Linck, John Dennis et al. — Monde

Source : arXiv

Résumé

Legacy Fortran physics packages in atmospheric models such as CESM, WRF, and E3SM pose a barrier to GPU modernization, while manual porting requires accelerator-programming expertise. We present LAFT (LLM-Assisted Fortran Translation), an agentic framework for converting Fortran atmospheric physics parameterizations into GPU-accelerated JAX while preserving numerical agreement with the Fortran reference. LAFT coordinates source-code analysis, translation, repair, runtime validation, and performance profiling. It combines a deterministic front end for metadata extraction and bridge generation,…

Actions

Citation