Adrian Bulat

@adrian-bulat · 1 works

Researcher developing efficient vision transformers using recurrent single-block architectures.

Recent posts by Adrian Bulat

• One Block, Multiple Depths: Recurrent Vision Transformers with Depth-Programmed Experts - In this work, we show that a single Transformer block, applied recurrently, can match the accuracy of a full-depth vision encoder at comparable inference FLOPs without intermediate feature… · Paper · arxiv.org