A learned task register per task, instead of one global pruning rule, coordinates token selection, depth-wise budget allocation, and dense recovery, yielding 47.0 mIoU and 53.7 box AP at 1.3 times encoder speedup on ADE20K and COCO.
International Conference on Learning Representations (ICLR) , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Putting Registers to Work: Task Registers for Token Pruning in Vision Transformers
A learned task register per task, instead of one global pruning rule, coordinates token selection, depth-wise budget allocation, and dense recovery, yielding 47.0 mIoU and 53.7 box AP at 1.3 times encoder speedup on ADE20K and COCO.