Author

Thomas Deixelberger

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Preprint Sep 2026

The Domain Is a Residue: Adapting Self-Supervised Features, Not Generators

The RFA has about 160 times fewer trainable parameters than CycleGAN-Turbo and under a fifth of its per-condition training time, and carries weather, lighting and rendering style as a residue as a residue of 13 to 14% of the feature norm.

Thomas Deixelberger, Markus Steinberger · 0 citations