# Highlight Every Step: Knowledge Distillation via Collaborative Teaching

arXiv: https://arxiv.org/abs/1907.09643 · HTML: https://mlsanity.com/paper/1907.09643/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jul 2019.

## Built on by (most cited first, 21 of 21)

- [AM-RADIO: Agglomerative Vision Foundation Model -- Reduce All Domains Into One](https://mlsanity.com/paper/2312.06709/) (2023)
- [A Machine Learning-oriented Survey on Tiny Machine Learning](https://mlsanity.com/paper/2309.11932/) (2023)
- [Single-Layer Vision Transformers for More Accurate Early Exits with Less Overhead](https://mlsanity.com/paper/2105.09121/) (2021)
- [PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation](https://mlsanity.com/paper/2410.01680/) (2024)
- [Mitigating Accuracy-Robustness Trade-off via Balanced Multi-Teacher Adversarial Distillation](https://mlsanity.com/paper/2306.16170/) (2023)
- [A Comprehensive Survey on Imbalanced Data Learning](https://mlsanity.com/paper/2502.08960/) (2025)
- [Dual Discriminator Adversarial Distillation for Data-free Model Compression](https://mlsanity.com/paper/2104.05382/) (2021)
- [Knowledge Distillation from Multiple Foundation Models for End-to-End Speech Recognition](https://mlsanity.com/paper/2303.10917/) (2023)
- [Designing and Training of Lightweight Neural Networks on Edge Devices using Early Halting in Knowledge Distillation](https://mlsanity.com/paper/2209.15560/) (2022)
- [DynamicKD: An Effective Knowledge Distillation via Dynamic Entropy Correction-Based Distillation for Gap Optimizing](https://mlsanity.com/paper/2305.05233/) (2023)
- [Improve Knowledge Distillation via Label Revision and Data Selection](https://mlsanity.com/paper/2404.03693/) (2024)
- [A Closer Look at Branch Classifiers of Multi-exit Architectures](https://mlsanity.com/paper/2204.13347/) (2022)
- [T3DNet: Compressing Point Cloud Models for Lightweight 3D Recognition](https://mlsanity.com/paper/2402.19264/) (2024)
- [Heterogeneous Generative Knowledge Distillation with Masked Image Modeling](https://mlsanity.com/paper/2309.09571/) (2023)
- [Knowledge Distillation: Enhancing Neural Network Compression with Integrated Gradients](https://mlsanity.com/paper/2503.13008/) (2025)
- [Instance-aware Model Ensemble With Distillation For Unsupervised Domain Adaptation](https://mlsanity.com/paper/2211.08106/) (2022)
- [The Staged Knowledge Distillation in Video Classification: Harmonizing Student Progress by a Complementary Weakly Supervised Framework](https://mlsanity.com/paper/2307.05201/) (2023)
- [Gap Preserving Distillation by Building Bidirectional Mappings with A Dynamic Teacher](https://mlsanity.com/paper/2410.04140/) (2024)
- [Model compression using knowledge distillation with integrated gradients](https://mlsanity.com/paper/2506.14440/) (2025)
- [Rank-Aware Agglomeration of Foundation Models for Immunohistochemistry Image Cell Counting](https://mlsanity.com/paper/2511.12588/) (2025)
- [Label-Guided Knowledge Distillation for 3D-CNNs in Action Recognition](https://mlsanity.com/paper/2609.13024/) (2026)

## Abstract (arXiv)

High storage and computational costs obstruct deep neural networks to be deployed on resource-constrained devices. Knowledge distillation aims to train a compact student network by transferring knowledge from a larger pre-trained teacher model. However, most existing methods on knowledge distillation ignore the valuable information among training process associated with training results. In this paper, we provide a new Collaborative Teaching Knowledge Distillation (CTKD) strategy which employs two special teachers. Specifically, one teacher trained from scratch (i.e., scratch teacher) assists the student step by step using its temporary outputs. It forces the student to approach the optimal path towards the final logits with high accuracy. The other pre-trained teacher (i.e., expert teacher) guides the student to focus on a critical region which is more useful for the task. The combination of the knowledge from two special teachers can significantly improve the performance of the student network in knowledge distillation. The results of experiments on CIFAR-10, CIFAR-100, SVHN and Tiny ImageNet datasets verify that the proposed knowledge distillation method is efficient and achieves state-of-the-art performance.