Open Models Almanac
Methodspost-trainingsupervised fine-tuning

single mention · post-training

Freeze language and visual backbones during fine-tuning

Keeps the language and visual backbones fixed while training other modules.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See supervised fine-tuning for the methods filed beside it.

Also called freezes.

Filed under post-training › supervised fine-tuning