Open Models Almanac
Methodsmodel architecturemultimodal architecture

single mention · model architecture

Shared-sequence multimodal input projection

Projects visual features, robot state, and language tokens into embeddings and concatenates them along the sequence dimension.

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 a flexible input processing pipeline.

Filed under model architecture › multimodal architecture