MixTok: Adaptive Visual Tokenization with Global 1D Tokens and Sparse 2D Refinement
MixTok mixes compact 1D tokens with sparse 2D refinement for budgeted visual tokenization.
1 stories, newest first · one card per story
MixTok mixes compact 1D tokens with sparse 2D refinement for budgeted visual tokenization.
Today 1.6K readers · 2.4K views · 774 reads · all time 189K views