https://hgpu.org/?p=13437
Extending the Gotran framework: LATEX and GPU acceleration