# Revisiting Self-Distillation

arXiv: https://arxiv.org/abs/2206.08491 · HTML: https://mlsanity.com/paper/2206.08491/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation.
- Builds on 46 of its 66 references that are in the map.
- First posted on arXiv 17 Jun 2022.

## Placed in

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

## Builds on (most cited first, 30 of 46)

- [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)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Improved Regularization of Convolutional Neural Networks with Cutout](https://mlsanity.com/paper/1708.04552/) (2017)
- [Knowledge Distillation: A Survey](https://mlsanity.com/paper/2006.05525/) (2020)
- [Sharpness-Aware Minimization for Efficiently Improving Generalization](https://mlsanity.com/paper/2010.01412/) (2020)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima](https://mlsanity.com/paper/1609.04836/) (2016)
- [Visualizing the Loss Landscape of Neural Nets](https://mlsanity.com/paper/1712.09913/) (2017)
- [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)
- [AutoAugment: Learning Augmentation Policies from Data](https://mlsanity.com/paper/1805.09501/) (2018)
- [Ensemble Distillation for Robust Model Fusion in Federated Learning](https://mlsanity.com/paper/2006.07242/) (2020)
- [Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data](https://mlsanity.com/paper/1703.11008/) (2017)
- [Fantastic Generalization Measures and Where to Find Them](https://mlsanity.com/paper/1912.02178/) (2019)
- [Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation](https://mlsanity.com/paper/1905.08094/) (2019)
- [Sharp Minima Can Generalize For Deep Nets](https://mlsanity.com/paper/1703.04933/) (2017)
- [Dreaming to Distill: Data-free Knowledge Transfer via DeepInversion](https://mlsanity.com/paper/1912.08795/) (2019)
- [On the Efficacy of Knowledge Distillation](https://mlsanity.com/paper/1910.01348/) (2019)
- [Entropy-SGD: Biasing Gradient Descent Into Wide Valleys](https://mlsanity.com/paper/1611.01838/) (2016)
- [Variational Information Distillation for Knowledge Transfer](https://mlsanity.com/paper/1904.05835/) (2019)
- [Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning](https://mlsanity.com/paper/2012.09816/) (2020)

## Built on by (most cited first, 24 of 24)

- [Understanding Self-Distillation in the Presence of Label Noise](https://mlsanity.com/paper/2301.13304/) (2023)
- [PRIME: Large Language Model Personalization with Cognitive Dual-Memory and Personalized Thought Process](https://mlsanity.com/paper/2507.04607/) (2025)
- [On student-teacher deviations in distillation: does it pay to disobey?](https://mlsanity.com/paper/2301.12923/) (2023)
- [Towards a theory of model distillation](https://mlsanity.com/paper/2403.09053/) (2024)
- [Align and Distill: Unifying and Improving Domain Adaptive Object Detection](https://mlsanity.com/paper/2403.12029/) (2024)
- [Entropy-Preserving Supervised Fine-Tuning via Adaptive Self-Distillation for Large Reasoning Models](https://mlsanity.com/paper/2602.02244/) (2026)
- [Fast Yet Effective Speech Emotion Recognition with Self-distillation](https://mlsanity.com/paper/2210.14636/) (2022)
- [Beyond Invariance: Test-Time Label-Shift Adaptation for Distributions with "Spurious" Correlations](https://mlsanity.com/paper/2211.15646/) (2022)
- [Simulated Annealing in Early Layers Leads to Better Generalization](https://mlsanity.com/paper/2304.04858/) (2023)
- [SDRT: Enhance Vision-Language Models by Self-Distillation with Diverse Reasoning Traces](https://mlsanity.com/paper/2503.01754/) (2025)
- [FedDistill: Global Model Distillation for Local Model De-Biasing in Non-IID Federated Learning](https://mlsanity.com/paper/2404.09210/) (2024)
- [Quantifying Context Bias in Domain Adaptation for Object Detection](https://mlsanity.com/paper/2409.14679/) (2024)
- [ATAS: Any-to-Any Self-Distillation for Enhanced Open-Vocabulary Dense Prediction](https://mlsanity.com/paper/2506.08678/) (2025)
- [SRPO: Self-Reflective Policy Optimization for Long-Horizon Reasoning](https://mlsanity.com/paper/2608.23493/) (2026)
- [Smooth-Distill: A Self-distillation Framework for Multitask Learning with Wearable Sensor Data](https://mlsanity.com/paper/2507.00061/) (2025)
- [On the Generalization of Knowledge Distillation: An Information-Theoretic View](https://mlsanity.com/paper/2605.13143/) (2026)
- [SADT: Combining Sharpness-Aware Minimization with Self-Distillation for Improved Model Generalization](https://mlsanity.com/paper/2211.00310/) (2022)
- [UNFUSED: UNsupervised Finetuning Using SElf supervised Distillation](https://mlsanity.com/paper/2303.05668/) (2023)
- [Bayesian Optimization Meets Self-Distillation](https://mlsanity.com/paper/2304.12666/) (2023)
- [An Efficient Self-Learning Framework For Interactive Spoken Dialog Systems](https://mlsanity.com/paper/2409.10515/) (2024)
- [Mitigating Context Bias in Domain Adaptation for Object Detection using Mask Pooling](https://mlsanity.com/paper/2505.18446/) (2025)
- [HAD: Heterogeneity-Aware Distillation for Lifelong Heterogeneous Learning](https://mlsanity.com/paper/2603.26192/) (2026)
- [Generative Diffusion Prior Distillation for Long-Context Knowledge Transfer](https://mlsanity.com/paper/2605.11414/) (2026)
- [Misanthrope: A Privacy-Preserving Keypoint Detector](https://mlsanity.com/paper/2608.23012/) (2026)

## Abstract (arXiv)

Knowledge distillation is the procedure of transferring "knowledge" from a large model (the teacher) to a more compact one (the student), often being used in the context of model compression. When both models have the same architecture, this procedure is called self-distillation. Several works have anecdotally shown that a self-distilled student can outperform the teacher on held-out data. In this work, we systematically study self-distillation in a number of settings. We first show that even with a highly accurate teacher, self-distillation allows a student to surpass the teacher in all cases. Secondly, we revisit existing theoretical explanations of (self) distillation and identify contradicting examples, revealing possible drawbacks of these explanations. Finally, we provide an alternative explanation for the dynamics of self-distillation through the lens of loss landscape geometry. We conduct extensive experiments to show that self-distillation leads to flatter minima, thereby resulting in better generalization.