نسخة أولية وصول مفتوح
GTR: Gated Token Recurrence for Efficient Dense Prediction
Self-attention-based vision backbones perform well on dense prediction, but the quadratic computational cost of global softmax attention limits their efficiency as image resolution increases. We introduce Gated Token Recurrence (GTR), a softmax-free recurrent vision backbone that combines gated linear attention, altern …