# m2mKD: Module-to-Module Knowledge Distillation for Modular Transformers

arXiv: https://arxiv.org/abs/2402.16918 · HTML: https://mlsanity.com/paper/2402.16918/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Expert routing strategies, one of 12 directions in Mixture-of-experts.
- Builds on 36 of its 45 references that are in the map.
- First posted on arXiv 26 Feb 2024.

## Placed in

- Mixture-of-experts / [Expert routing strategies](https://mlsanity.com/moe/expert-routing-strategies/)
- Knowledge distillation / [General knowledge distillation methods](https://mlsanity.com/distill/general-knowledge-distillation-methods/)

## Builds on (most cited first, 30 of 36)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Scaling Vision with Sparse Mixture of Experts](https://mlsanity.com/paper/2106.05974/) (2021)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [Improved Knowledge Distillation via Teacher Assistant](https://mlsanity.com/paper/1902.03393/) (2019)
- [Do Deep Nets Really Need to be Deep?](https://mlsanity.com/paper/1312.6184/) (2013)
- [Dynamic Neural Networks: A Survey](https://mlsanity.com/paper/2102.04906/) (2021)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [Paraphrasing Complex Network: Network Compression via Factor Transfer](https://mlsanity.com/paper/1802.04977/) (2018)
- [Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints](https://mlsanity.com/paper/2212.05055/) (2022)
- [Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Experts](https://mlsanity.com/paper/2206.02770/) (2022)
- [Tutel: Adaptive Mixture-of-Experts at Scale](https://mlsanity.com/paper/2206.03382/) (2022)
- [Efficient parametrization of multi-domain deep neural networks](https://mlsanity.com/paper/1803.10082/) (2018)
- [MoEfication: Transformer Feed-forward Layers are Mixtures of Experts](https://mlsanity.com/paper/2110.01786/) (2021)
- [DEMix Layers: Disentangling Domains for Modular Language Modeling](https://mlsanity.com/paper/2108.05036/) (2021)
- [BERT-of-Theseus: Compressing BERT by Progressive Module Replacing](https://mlsanity.com/paper/2002.02925/) (2020)
- [Coordination Among Neural Modules Through a Shared Global Workspace](https://mlsanity.com/paper/2103.01197/) (2021)
- [Densely Guided Knowledge Distillation using Multiple Teacher Assistants](https://mlsanity.com/paper/2009.08825/) (2020)
- [Modular Deep Learning](https://mlsanity.com/paper/2302.11529/) (2023)
- [Deep Model Reassembly](https://mlsanity.com/paper/2210.17409/) (2022)

## Built on by (most cited first, 5 of 5)

- [Efficient Multimodal Large Language Models: A Survey](https://mlsanity.com/paper/2405.10739/) (2024)
- [From Efficient Multimodal Models to World Models: A Survey](https://mlsanity.com/paper/2407.00118/) (2024)
- [Embedding Domain Knowledge for Large Language Models via Reinforcement Learning from Augmented Generation](https://mlsanity.com/paper/2509.20162/) (2025)
- [Locking Pretrained Weights via Deep Low-Rank Residual Distillation](https://mlsanity.com/paper/2605.10777/) (2026)
- [Mixture of Training: Recombining Small-Scale Scaffolded Pretraining Runs into a Larger Language Model](https://mlsanity.com/paper/2608.13277/) (2026)

## Abstract (arXiv)

Modular neural architectures are gaining attention for their powerful generalization and efficient adaptation to new domains. However, training these models poses challenges due to optimization difficulties arising from intrinsic sparse connectivity. Leveraging knowledge from monolithic models through techniques like knowledge distillation can facilitate training and enable integration of diverse knowledge. Nevertheless, conventional knowledge distillation approaches are not tailored to modular models and struggle with unique architectures and enormous parameter counts. Motivated by these challenges, we propose module-to-module knowledge distillation (m2mKD) for transferring knowledge between modules. m2mKD combines teacher modules of a pretrained monolithic model and student modules of a modular model with a shared meta model respectively to encourage the student module to mimic the behaviour of the teacher module. We evaluate m2mKD on two modular neural architectures: Neural Attentive Circuits (NACs) and Vision Mixture-of-Experts (V-MoE). Applying m2mKD to NACs yields significant improvements in IID accuracy on Tiny-ImageNet (up to 5.6%) and OOD robustness on Tiny-ImageNet-R (up to 4.2%). Additionally, the V-MoE-Base model trained with m2mKD achieves 3.5% higher accuracy than end-to-end training on ImageNet-1k. Code is available at https://github.com/kamanphoebe/m2mKD.