# Co-training 2L Submodels for Visual Recognition

arXiv: https://arxiv.org/abs/2212.04884 · HTML: https://mlsanity.com/paper/2212.04884/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation.
- Builds on 44 of its 65 references that are in the map.
- First posted on arXiv 9 Dec 2022.

## Placed in

- Knowledge distillation / [Detection and dense prediction distillation](https://mlsanity.com/distill/detection-and-dense-prediction-distillation/)

## Builds on (most cited first, 30 of 44)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [EfficientNetV2: Smaller Models and Faster Training](https://mlsanity.com/paper/2104.00298/) (2021)
- [Deep Networks with Stochastic Depth](https://mlsanity.com/paper/1603.09382/) (2016)
- [Designing Network Design Spaces](https://mlsanity.com/paper/2003.13678/) (2020)
- [Going deeper with Image Transformers](https://mlsanity.com/paper/2103.17239/) (2021)
- [Deep Mutual Learning](https://mlsanity.com/paper/1706.00384/) (2017)
- [CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows](https://mlsanity.com/paper/2107.00652/) (2021)
- [Do Deep Nets Really Need to be Deep?](https://mlsanity.com/paper/1312.6184/) (2013)
- [Reducing Transformer Depth on Demand with Structured Dropout](https://mlsanity.com/paper/1909.11556/) (2019)
- [DeiT III: Revenge of the ViT](https://mlsanity.com/paper/2204.07118/) (2022)
- [ResNet strikes back: An improved training procedure in timm](https://mlsanity.com/paper/2110.00476/) (2021)

## Built on by (most cited first, 8 of 8)

- [Vision-LSTM: xLSTM as Generic Vision Backbone](https://mlsanity.com/paper/2406.04303/) (2024)
- [CR-CTC: Consistency regularization on CTC for improved speech recognition](https://mlsanity.com/paper/2410.05101/) (2024)
- [Morphing Tokens Draw Strong Masked Image Models](https://mlsanity.com/paper/2401.00254/) (2023)
- [Learning with Unmasked Tokens Drives Stronger Vision Learners](https://mlsanity.com/paper/2310.13593/) (2023)
- [MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation](https://mlsanity.com/paper/2411.19067/) (2024)
- [Adaptive Depth Networks with Skippable Sub-Paths](https://mlsanity.com/paper/2312.16392/) (2023)
- [POA: Pre-training Once for Models of All Sizes](https://mlsanity.com/paper/2408.01031/) (2024)
- [Improving Audio Event Recognition with Consistency Regularization](https://mlsanity.com/paper/2509.10391/) (2025)

## Abstract (arXiv)

We introduce submodel co-training, a regularization method related to co-training, self-distillation and stochastic depth. Given a neural network to be trained, for each sample we implicitly instantiate two altered networks, ``submodels'', with stochastic depth: we activate only a subset of the layers. Each network serves as a soft teacher to the other, by providing a loss that complements the regular loss provided by the one-hot label. Our approach, dubbed cosub, uses a single set of weights, and does not involve a pre-trained external model or temporal averaging. Experimentally, we show that submodel co-training is effective to train backbones for recognition tasks such as image classification and semantic segmentation. Our approach is compatible with multiple architectures, including RegNet, ViT, PiT, XCiT, Swin and ConvNext. Our training strategy improves their results in comparable settings. For instance, a ViT-B pretrained with cosub on ImageNet-21k obtains 87.4% top-1 acc. @448 on ImageNet-val.