# FerKD: Surgical Label Adaptation for Efficient Distillation

arXiv: https://arxiv.org/abs/2312.17473 · HTML: https://mlsanity.com/paper/2312.17473/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation.
- Builds on 45 of its 58 references that are in the map.
- First posted on arXiv 29 Dec 2023.

## Placed in

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

## Builds on (most cited first, 30 of 45)

- [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)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [When Does Label Smoothing Help?](https://mlsanity.com/paper/1906.02629/) (2019)
- [Relational Knowledge Distillation](https://mlsanity.com/paper/1904.05068/) (2019)
- [EVA: Exploring the Limits of Masked Visual Representation Learning at Scale](https://mlsanity.com/paper/2211.07636/) (2022)
- [Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks](https://mlsanity.com/paper/1511.04508/) (2015)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers](https://mlsanity.com/paper/2106.10270/) (2021)
- [Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation](https://mlsanity.com/paper/1905.08094/) (2019)
- [ResNet strikes back: An improved training procedure in timm](https://mlsanity.com/paper/2110.00476/) (2021)
- [Are we done with ImageNet?](https://mlsanity.com/paper/2006.07159/) (2020)
- [Structured Knowledge Distillation for Dense Prediction](https://mlsanity.com/paper/1903.04197/) (2019)
- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [Knowledge distillation: A good teacher is patient and consistent](https://mlsanity.com/paper/2106.05237/) (2021)
- [TinyViT: Fast Pretraining Distillation for Small Vision Transformers](https://mlsanity.com/paper/2207.10666/) (2022)
- [Does Knowledge Distillation Really Work?](https://mlsanity.com/paper/2106.05945/) (2021)
- [Distilling Object Detectors with Fine-grained Feature Imitation](https://mlsanity.com/paper/1906.03609/) (2019)
- [Better Fine-Tuning by Reducing Representational Collapse](https://mlsanity.com/paper/2008.03156/) (2020)
- [Focal and Global Knowledge Distillation for Detectors](https://mlsanity.com/paper/2111.11837/) (2021)

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

- [Are Large-scale Soft Labels Necessary for Large-scale Dataset Distillation?](https://mlsanity.com/paper/2410.15919/) (2024)
- [Active Data Curation Effectively Distills Large-Scale Multimodal Models](https://mlsanity.com/paper/2411.18674/) (2024)
- [FDBPL: Faster Distillation-Based Prompt Learning for Region-Aware Vision-Language Models Adaptation](https://mlsanity.com/paper/2505.18053/) (2025)
- [Vector-Quantized Soft Label Compression for Dataset Distillation](https://mlsanity.com/paper/2603.03808/) (2026)
- [Soft Label Pruning and Quantization for Large-Scale Dataset Distillation](https://mlsanity.com/paper/2604.18135/) (2026)

## Abstract (arXiv)

We present FerKD, a novel efficient knowledge distillation framework that incorporates partial soft-hard label adaptation coupled with a region-calibration mechanism. Our approach stems from the observation and intuition that standard data augmentations, such as RandomResizedCrop, tend to transform inputs into diverse conditions: easy positives, hard positives, or hard negatives. In traditional distillation frameworks, these transformed samples are utilized equally through their predictive probabilities derived from pretrained teacher models. However, merely relying on prediction values from a pretrained teacher, a common practice in prior studies, neglects the reliability of these soft label predictions. To address this, we propose a new scheme that calibrates the less-confident regions to be the context using softened hard groundtruth labels. Our approach involves the processes of hard regions mining + calibration. We demonstrate empirically that this method can dramatically improve the convergence speed and final accuracy. Additionally, we find that a consistent mixing strategy can stabilize the distributions of soft supervision, taking advantage of the soft labels. As a result, we introduce a stabilized SelfMix augmentation that weakens the variation of the mixed images and corresponding soft labels through mixing similar regions within the same image. FerKD is an intuitive and well-designed learning system that eliminates several heuristics and hyperparameters in former FKD solution. More importantly, it achieves remarkable improvement on ImageNet-1K and downstream tasks. For instance, FerKD achieves 81.2% on ImageNet-1K with ResNet-50, outperforming FKD and FunMatch by remarkable margins. Leveraging better pre-trained weights and larger architectures, our finetuned ViT-G14 even achieves 89.9%. Our code is available at https://github.com/szq0214/FKD/tree/main/FerKD.