implementation detail · filed under model architecture
Lightweight vision embedding module
A vision embedding module composed of a matrix multiplication, positional embedding, and normalizations.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
We replaced Gemma 4’s vision encoder with a lightweight embedding module consisting of a single matrix multiplication, positional embedding and normalizations.
usedmodel architecturein Gemma 4 12BGoogle
Filed alongside
Other methods under model architecture :: multimodal architecture.
Encoder-free multimodal architectureEarly fusion multimodal trainingDiscrete token encoding for audioHierarchical patch encoderMixed-modality trainingConfigurable visual token budgetJoint multimodal decoding in a shared hidden spaceNative multimodal understanding3×3 pixel-unshuffle downsamplingJoint vision-language pre-trainingRaw audio projection into the LLM embedding spaceVision Transformer (ViT)2×2 pixel-shuffle downsamplingAudio encoder initialized from MiMo-AudioAudio Transformer (AuT)Causal streaming ConvNet codec decoderData-parallel-first multimodal encodingDecoupled Encoder Process (DEP)Dedicated multimodal encodersDisaggregated encoder trainingDiscarding the LLM after vision-encoder trainingDual-format coordinate supervisionExplicit text-string timestampsFour-frame audio patches with within-patch bidirectional self-attention