# MoE Jetpack: From Dense Checkpoints to Adaptive Mixture of Experts for Vision Tasks

arXiv: https://arxiv.org/abs/2406.04801 · HTML: https://mlsanity.com/paper/2406.04801/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts.
- Builds on 43 of its 57 references that are in the map.
- First posted on arXiv 7 Jun 2024.

## Placed in

- Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/)

## Builds on (most cited first, 30 of 43)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [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)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Random Erasing Data Augmentation](https://mlsanity.com/paper/1708.04896/) (2017)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models](https://mlsanity.com/paper/2401.06066/) (2024)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [Going deeper with Image Transformers](https://mlsanity.com/paper/2103.17239/) (2021)
- [Scaling Vision with Sparse Mixture of Experts](https://mlsanity.com/paper/2106.05974/) (2021)
- [Mixture-of-Experts with Expert Choice Routing](https://mlsanity.com/paper/2202.09368/) (2022)
- [Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning](https://mlsanity.com/paper/2310.06694/) (2023)

## Built on by (most cited first, 9 of 9)

- [TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters](https://mlsanity.com/paper/2410.23168/) (2024)
- [Upcycling Instruction Tuning from Dense to Mixture-of-Experts via Parameter Merging](https://mlsanity.com/paper/2410.01610/) (2024)
- [SToLa: Self-Adaptive Touch-Language Framework with Tactile Commonsense Reasoning in Open-Ended Scenarios](https://mlsanity.com/paper/2505.04201/) (2025)
- [Mamba Goes HoME: Hierarchical Soft Mixture-of-Experts for 3D Medical Image Segmentation](https://mlsanity.com/paper/2507.06363/) (2025)
- [Mixture-of-Experts for Personalized and Semantic-Aware Next Location Prediction](https://mlsanity.com/paper/2505.24597/) (2025)
- [UME: Upcycling Mixture-of-Experts for Scalable and Efficient Automatic Speech Recognition](https://mlsanity.com/paper/2412.17507/) (2024)
- [Mixture of Group Experts for Learning Invariant Representations](https://mlsanity.com/paper/2504.09265/) (2025)
- [R2M: Real-Aware Residual Model Merging for Robust and Generalizable Deepfake Detection](https://mlsanity.com/paper/2509.24367/) (2025)
- [Improving Ensemble Forecasts of Abnormally Deflecting Tropical Cyclones with Fused Atmosphere-Ocean-Terrain Data](https://mlsanity.com/paper/2603.29200/) (2026)

## Abstract (arXiv)

The sparsely activated mixture of experts (MoE) model presents a promising alternative to traditional densely activated (dense) models, enhancing both quality and computational efficiency. However, training MoE models from scratch demands extensive data and computational resources. Moreover, public repositories like timm mainly provide pre-trained dense checkpoints, lacking similar resources for MoE models, hindering their adoption. To bridge this gap, we introduce MoE Jetpack, an effective method for fine-tuning dense checkpoints into MoE models. MoE Jetpack incorporates two key techniques: (1) checkpoint recycling, which repurposes dense checkpoints as initial weights for MoE models, thereby accelerating convergence, enhancing accuracy, and alleviating the computational burden of pre-training; (2) hyperspherical adaptive MoE (SpheroMoE) layer, which optimizes the MoE architecture for better integration of dense checkpoints, enhancing fine-tuning performance. Our experiments on vision tasks demonstrate that MoE Jetpack significantly improves convergence speed and accuracy when fine-tuning dense checkpoints into MoE models. Our code will be publicly available at https://github.com/Adlith/MoE-Jetpack.