Open Models Almanac
Methodsmodel architecturemultimodal architecture

single mention · model architecture

Three-stage multimodal training protocol

Trains the multimodal model in three stages, introducing multimodal alignment at selected points.

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

Also called three stages.

Filed under model architecture › multimodal architecture