Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces reViT, a recurrent Vision Transformer that uses a single Transformer block applied multiple times to achieve performance comparable to deeper encoders without intermediate distillation. It achieves this by representing the FFN at each recurrent depth as a convex combination of shared experts, programmed by a continuous normalized-depth coordinate. This approach yields significant parameter reduction and strong performance across supervised training and distillation, even enabling elastic-depth operation from a single checkpoint.
A single recurrent Transformer block can match the accuracy of deep vision encoders with 70% fewer parameters by dynamically programming FFN experts via a continuous depth coordinate.
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 distillation. reViT restores depth-specific transformations by representing the FFN at each recurrent depth as a convex combination of a small shared expert bank. A continuous normalized-depth coordinate programs this mixture, defining a resampleable trajectory through FFN parameter space. We evaluate this design in two regimes: supervised ImageNet-1k training and distillation from a DINOv2 teacher. Across both regimes, controlled adaptations identify weight-space merging as the strongest tested MoE family at a matching one-FFN budget, ahead of the token-dispatch and output-mixture alternatives. Trained from scratch, reViT-B/16 attains DeiT III accuracy with about 70\% fewer stored parameters. An 8-experts model distilled using only the teacher's output features retains nearly all of its DINOv2 teacher's linear-probe accuracy and transfers across classification, segmentation, and depth prediction. Elastic-depth training allows one checkpoint (trained model) to operate at multiple tested depths by resampling the same normalized coordinate interval. For fixed-depth deployment, the recurrent block can be materialized as a conventional dense graph, removing online routing and merging without changing the one-FFN-per-depth compute but expanding deployment storage.