# Improving Knowledge Distillation via Regularizing Feature Norm and Direction

arXiv: https://arxiv.org/abs/2305.17007 · HTML: https://mlsanity.com/paper/2305.17007/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation.
- Builds on 30 of its 40 references that are in the map.
- First posted on arXiv 26 May 2023.

## Placed in

- Knowledge distillation / [General knowledge distillation methods](https://mlsanity.com/distill/general-knowledge-distillation-methods/)

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

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [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)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices](https://mlsanity.com/paper/1707.01083/) (2017)
- [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://mlsanity.com/paper/1807.11164/) (2018)
- [Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer](https://mlsanity.com/paper/1612.03928/) (2016)
- [Relational Knowledge Distillation](https://mlsanity.com/paper/1904.05068/) (2019)
- [Contrastive Representation Distillation](https://mlsanity.com/paper/1910.10699/) (2019)
- [DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition](https://mlsanity.com/paper/1310.1531/) (2013)
- [Decoupled Knowledge Distillation](https://mlsanity.com/paper/2203.08679/) (2022)
- [A Comprehensive Overhaul of Feature Distillation](https://mlsanity.com/paper/1904.01866/) (2019)
- [CNN Features off-the-shelf: an Astounding Baseline for Recognition](https://mlsanity.com/paper/1403.6382/) (2014)
- [Distilling Knowledge via Knowledge Review](https://mlsanity.com/paper/2104.09044/) (2021)
- [Knowledge distillation: A good teacher is patient and consistent](https://mlsanity.com/paper/2106.05237/) (2021)
- [Learning Deep Representations with Probabilistic Knowledge Transfer](https://mlsanity.com/paper/1803.10837/) (2018)
- [Knowledge Distillation from A Stronger Teacher](https://mlsanity.com/paper/2205.10536/) (2022)
- [Distilling Object Detectors with Fine-grained Feature Imitation](https://mlsanity.com/paper/1906.03609/) (2019)
- [OpenGAN: Open-Set Recognition via Open Data Generation](https://mlsanity.com/paper/2104.02939/) (2021)
- [Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers](https://mlsanity.com/paper/1802.00124/) (2018)

## Built on by (most cited first, 18 of 18)

- [Knowledge Distillation Based on Transformed Teacher Matching](https://mlsanity.com/paper/2402.11148/) (2024)
- [VLM-KD: Knowledge Distillation from VLM for Long-Tail Visual Recognition](https://mlsanity.com/paper/2408.16930/) (2024)
- [Feature Alignment and Representation Transfer in Knowledge Distillation for Large Language Models](https://mlsanity.com/paper/2504.13825/) (2025)
- [Perspective-Aware Teaching: Adapting Knowledge for Heterogeneous Distillation](https://mlsanity.com/paper/2501.08885/) (2025)
- [Dual-Head Knowledge Distillation: Enhancing Logits Utilization with an Auxiliary Head](https://mlsanity.com/paper/2411.08937/) (2024)
- [Cross-Architecture Distillation Made Simple with Redundancy Suppression](https://mlsanity.com/paper/2507.21844/) (2025)
- [One-Shot Knowledge Transfer for Scalable Person Re-Identification](https://mlsanity.com/paper/2511.06016/) (2025)
- [Progressive Feature Self-reinforcement for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2312.08916/) (2023)
- [FedHPL: Efficient Heterogeneous Federated Learning with Prompt Tuning and Logit Distillation](https://mlsanity.com/paper/2405.17267/) (2024)
- [Neural Collapse Inspired Knowledge Distillation](https://mlsanity.com/paper/2412.11788/) (2024)
- [Proxy-FDA: Proxy-based Feature Distribution Alignment for Fine-tuning Vision Foundation Models without Forgetting](https://mlsanity.com/paper/2505.24088/) (2025)
- [Ever Evolving Evaluator (EV3): Towards Flexible and Reliable Meta-Optimization for Knowledge Distillation](https://mlsanity.com/paper/2310.18893/) (2023)
- [Unveiling Incomplete Modality Brain Tumor Segmentation: Leveraging Masked Predicted Auto-Encoder and Divergence Learning](https://mlsanity.com/paper/2406.08634/) (2024)
- [A Comprehensive Study of Structural Pruning for Vision Models](https://mlsanity.com/paper/2406.12315/) (2024)
- [Unified Framework for Pre-trained Neural Network Compression via Decomposition and Optimized Rank Selection](https://mlsanity.com/paper/2409.03555/) (2024)
- [Knowledge Migration Framework for Smart Contract Vulnerability Detection](https://mlsanity.com/paper/2412.11175/) (2024)
- [VLCD: Vision-Language Contrastive Distillation for Accurate and Efficient Automatic Placenta Analysis](https://mlsanity.com/paper/2506.02229/) (2025)
- [Logit Distance Bounds Representational Similarity](https://mlsanity.com/paper/2602.15438/) (2026)

## Abstract (arXiv)

Knowledge distillation (KD) exploits a large well-trained model (i.e., teacher) to train a small student model on the same dataset for the same task. Treating teacher features as knowledge, prevailing methods of knowledge distillation train student by aligning its features with the teacher's, e.g., by minimizing the KL-divergence between their logits or L2 distance between their intermediate features. While it is natural to believe that better alignment of student features to the teacher better distills teacher knowledge, simply forcing this alignment does not directly contribute to the student's performance, e.g., classification accuracy. In this work, we propose to align student features with class-mean of teacher features, where class-mean naturally serves as a strong classifier. To this end, we explore baseline techniques such as adopting the cosine distance based loss to encourage the similarity between student features and their corresponding class-means of the teacher. Moreover, we train the student to produce large-norm features, inspired by other lines of work (e.g., model pruning and domain adaptation), which find the large-norm features to be more significant. Finally, we propose a rather simple loss term (dubbed ND loss) to simultaneously (1) encourage student to produce large-norm features, and (2) align the direction of student features and teacher class-means. Experiments on standard benchmarks demonstrate that our explored techniques help existing KD methods achieve better performance, i.e., higher classification accuracy on ImageNet and CIFAR100 datasets, and higher detection precision on COCO dataset. Importantly, our proposed ND loss helps the most, leading to the state-of-the-art performance on these benchmarks. The source code is available at urlhttps://github.com/WangYZ1608/Knowledge-Distillation-via-ND.