Open Models Almanac
Methodspost-trainingsupervised fine-tuning

single mention · post-training

Robot-only fine-tuning

Trains on robot data without a multimodal VLM mixture or separate VLM dataloader.

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 robot-only.

Filed under post-training › supervised fine-tuning