Open Models Almanac
Methodsmodel architecturemultimodal architecture

single mention · model architecture

PixelShuffle 2×2 visual-token downsampling

Uses a 2×2 pixel-shuffle operation to reduce the number of visual tokens by a factor of four.

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.

Filed under model architecture › multimodal architecture