# Reversible Vision Transformers

Source: [Reversible Vision Transformers](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4041c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:27:49.863Z

Authors: Karttikeya Mangalam, Haoqi Fan, Yanghao Li, Chao-Yuan Wu, Bo Xiong, Christoph Feichtenhofer, Jitendra Malik

Published: 2023-02-09T18:59:54.000Z

## Abstract

We present Reversible Vision Transformers, a memory efficient architecture design for visual recognition. By decoupling the GPU memory requirement from the depth of the model, Reversible Vision Transformers enable scaling up architectures with efficient memory usage. We adapt two popular models, namely Vision Transformer and Multiscale Vision Transformers, to reversible variants and benchmark extensively across both model sizes and tasks of image classification, object detection and video classification. Reversible Vision Transformers achieve a reduced memory footprint of up to 15.5x at roughly identical model complexity, parameters and accuracy, demonstrating the promise of reversible vision transformers as an efficient backbone for hardware resource limited training regimes. Finally, we find that the additional computational burden of recomputing activations is more than overcome for deeper models, where throughput can increase up to 2.3x over their non-reversible counterparts. Full code and trained models are available at https://github.com/facebookresearch/slowfast. A simpler, easy to understand and modify version is also available at https://github.com/karttikeya/minREV

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Video Classification on Kinetics-600 (test)](<https://www.sota2.com/research/sota/video-classification-on-kinetics-600-test.md>): Video Classification; Kinetics-600 (test); Top-1 Accuracy: 83.7.

- [Image Classification on ImageNet-1k](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k.md>): Image Classification; ImageNet-1K; Top-1 Acc: 82.5.

- [Action Recognition on Kinetics-400](<https://www.sota2.com/research/sota/action-recognition-on-kinetics-400.md>): Action Recognition; Kinetics-400; Top-1 Acc: 78.5.

- [Image Classification on ImageNet-1K 1.0 (val)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-1-0-val.md>): Image Classification; ImageNet-1K 1.0 (val); Top-1 Accuracy: 81.8.

- [Instance Segmentation on COCO 2017 (val)](<https://www.sota2.com/research/sota/instance-segmentation-on-coco-2017-val.md>): Instance Segmentation; COCO 2017 (val); APm: not recorded.

- [Object Detection on MS-COCO 2017 (val)](<https://www.sota2.com/research/sota/object-detection-on-ms-coco-2017-val.md>): Object Detection; MS-COCO 2017 (val); mAP: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2302.04869>)

[PDF](<https://arxiv.org/pdf/2302.04869.pdf>)

[Implementation](<https://github.com/facebookresearch/slowfast>)
