# From Knowledge Distillation to Self-Knowledge Distillation: A Unified Approach with Normalized Loss and Customized Soft Labels

arXiv: https://arxiv.org/abs/2303.13005 · HTML: https://mlsanity.com/paper/2303.13005/

## Key facts

- Cited by 69 papers in the map (data through 14 Oct 2026).
- Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation.
- Builds on 41 of its 55 references that are in the map.
- First posted on arXiv 23 Mar 2023.

## Placed in

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

## Builds on (most cited first, 30 of 41)

- [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)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [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)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [MMDetection: Open MMLab Detection Toolbox and Benchmark](https://mlsanity.com/paper/1906.07155/) (2019)
- [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)
- [When Does Label Smoothing Help?](https://mlsanity.com/paper/1906.02629/) (2019)
- [Designing Network Design Spaces](https://mlsanity.com/paper/2003.13678/) (2020)
- [Relational Knowledge Distillation](https://mlsanity.com/paper/1904.05068/) (2019)
- [Contrastive Representation Distillation](https://mlsanity.com/paper/1910.10699/) (2019)
- [Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation](https://mlsanity.com/paper/1905.08094/) (2019)
- [Decoupled Knowledge Distillation](https://mlsanity.com/paper/2203.08679/) (2022)
- [A Comprehensive Overhaul of Feature Distillation](https://mlsanity.com/paper/1904.01866/) (2019)
- [Distilling Knowledge via Knowledge Review](https://mlsanity.com/paper/2104.09044/) (2021)
- [Knowledge Distillation from A Stronger Teacher](https://mlsanity.com/paper/2205.10536/) (2022)
- [Channel-wise Knowledge Distillation for Dense Prediction](https://mlsanity.com/paper/2011.13256/) (2020)
- [Regularizing Class-wise Predictions via Self-knowledge Distillation](https://mlsanity.com/paper/2003.13964/) (2020)

## Built on by (most cited first, 30 of 69)

- [Effective Whole-body Pose Estimation with Two-stages Distillation](https://mlsanity.com/paper/2307.15880/) (2023)
- [What is the Role of Small Models in the LLM Era: A Survey](https://mlsanity.com/paper/2409.06857/) (2024)
- [Knowledge Distillation Based on Transformed Teacher Matching](https://mlsanity.com/paper/2402.11148/) (2024)
- [VkD: Improving Knowledge Distillation using Orthogonal Projections](https://mlsanity.com/paper/2403.06213/) (2024)
- [DDK: Distilling Domain Knowledge for Efficient Large Language Models](https://mlsanity.com/paper/2407.16154/) (2024)
- [BiLD: Bi-directional Logits Difference Loss for Large Language Model Distillation](https://mlsanity.com/paper/2406.13555/) (2024)
- [Wasserstein Distance Rivals Kullback-Leibler Divergence for Knowledge Distillation](https://mlsanity.com/paper/2412.08139/) (2024)
- [Rethinking Centered Kernel Alignment in Knowledge Distillation](https://mlsanity.com/paper/2401.11824/) (2024)
- [Weakly Supervised Monocular 3D Detection with a Single-View Image](https://mlsanity.com/paper/2402.19144/) (2024)
- [GIFT: Unlocking Full Potential of Labels in Distilled Dataset at Near-zero Cost](https://mlsanity.com/paper/2405.14736/) (2024)
- [Self-Distillation Learning Based on Temporal-Spatial Consistency for Spiking Neural Networks](https://mlsanity.com/paper/2406.07862/) (2024)
- [GIST: Improving Parameter Efficient Fine Tuning via Knowledge Interaction](https://mlsanity.com/paper/2312.07255/) (2023)
- [Direct Distillation between Different Domains](https://mlsanity.com/paper/2401.06826/) (2024)
- [Applications of Knowledge Distillation in Remote Sensing: A Survey](https://mlsanity.com/paper/2409.12111/) (2024)
- [CKD: Contrastive Knowledge Distillation from A Sample-wise Perspective](https://mlsanity.com/paper/2404.14109/) (2024)
- [Decoupling Dark Knowledge via Block-wise Logit Distillation for Feature-level Alignment](https://mlsanity.com/paper/2411.01547/) (2024)
- [MergeNet: Knowledge Migration across Heterogeneous Models, Tasks, and Modalities](https://mlsanity.com/paper/2404.13322/) (2024)
- [VRM: Knowledge Distillation via Virtual Relation Matching](https://mlsanity.com/paper/2502.20760/) (2025)
- [Scale Decoupled Distillation](https://mlsanity.com/paper/2403.13512/) (2024)
- [Low-Rank Knowledge Decomposition for Medical Foundation Models](https://mlsanity.com/paper/2404.17184/) (2024)
- [Cross-View Consistency Regularisation for Knowledge Distillation](https://mlsanity.com/paper/2412.16493/) (2024)
- [Unleashing the Potential of Mamba: Boosting a LiDAR 3D Sparse Detector by Using Cross-Model Knowledge Distillation](https://mlsanity.com/paper/2409.11018/) (2024)
- [Sample-level Adaptive Knowledge Distillation for Action Recognition](https://mlsanity.com/paper/2504.00606/) (2025)
- [DeepKD: A Deeply Decoupled and Denoised Knowledge Distillation Trainer](https://mlsanity.com/paper/2505.15133/) (2025)
- [Competitive Distillation: A Simple Learning Strategy for Improving Visual Classification](https://mlsanity.com/paper/2506.23285/) (2025)
- [Local Dense Logit Relations for Enhanced Knowledge Distillation](https://mlsanity.com/paper/2507.15911/) (2025)
- [Cross-Architecture Distillation Made Simple with Redundancy Suppression](https://mlsanity.com/paper/2507.21844/) (2025)
- [Positive-Unlabeled Reinforcement Learning Distillation for On-Premise Small Models](https://mlsanity.com/paper/2601.20687/) (2026)
- [Communication-Efficient Federated Knowledge Graph Embedding with Entity-Wise Top-K Sparsification](https://mlsanity.com/paper/2406.13225/) (2024)
- [Enhancing Logits Distillation with Plug&Play Kendall's τ Ranking Loss](https://mlsanity.com/paper/2409.17823/) (2024)

## Abstract (arXiv)

Knowledge Distillation (KD) uses the teacher's prediction logits as soft labels to guide the student, while self-KD does not need a real teacher to require the soft labels. This work unifies the formulations of the two tasks by decomposing and reorganizing the generic KD loss into a Normalized KD (NKD) loss and customized soft labels for both target class (image's category) and non-target classes named Universal Self-Knowledge Distillation (USKD). We decompose the KD loss and find the non-target loss from it forces the student's non-target logits to match the teacher's, but the sum of the two non-target logits is different, preventing them from being identical. NKD normalizes the non-target logits to equalize their sum. It can be generally used for KD and self-KD to better use the soft labels for distillation loss. USKD generates customized soft labels for both target and non-target classes without a teacher. It smooths the target logit of the student as the soft target label and uses the rank of the intermediate feature to generate the soft non-target labels with Zipf's law. For KD with teachers, our NKD achieves state-of-the-art performance on CIFAR-100 and ImageNet datasets, boosting the ImageNet Top-1 accuracy of ResNet18 from 69.90% to 71.96% with a ResNet-34 teacher. For self-KD without teachers, USKD is the first self-KD method that can be effectively applied to both CNN and ViT models with negligible additional time and memory cost, resulting in new state-of-the-art results, such as 1.17% and 0.55% accuracy gains on ImageNet for MobileNet and DeiT-Tiny, respectively. Our codes are available at https://github.com/yzd-v/cls_KD.