Open Models Almanac
Methodsdata curationdata mixture & curriculum

single mention · data curation

Co-training vision-language-action models with multimodal web data

Adding multimodal web data from a vision-language training mixture when co-training vision-language-action models.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See data mixture & curriculum for the methods filed beside it.

Filed under data curation › data mixture & curriculum