paper-with-me

Papers

Masked Generative Distillation

2022-05-03 · Zhendong Yang, Zhe Li, Mingqi Shao, Dachuan Shi, Zehuan Yuan, Chun Yuan

Knowledge distillation has been applied to various tasks successfully. The current distillation algorithm usually improves students' performance by imitating the output of the teacher. This paper shows that teachers can also improve students' representation power by guiding students' feature recovery. From this point of view, we propose Masked Generative Distillation (MGD), which is simple: we mask random pixels of the student's feature and force it to generate the teacher's full feature through a simple block. MGD is a truly general feature-based distillation method, which can be utilized on various tasks, including image classification, object detection, semantic segmentation and instance segmentation. We experiment on different models with extensive datasets and the results show that all the students achieve excellent improvements. Notably, we boost ResNet-18 from 69.90% to 71.69% ImageNet top-1 accuracy, RetinaNet with ResNet-50 backbone from 37.4 to 41.0 Boundingbox mAP, SOLO based on ResNet-50 from 33.1 to 36.2 Mask mAP and DeepLabV3 based on ResNet-18 from 73.20 to 76.02 mIoU. Our codes are available at https://github.com/yzd-v/MGD.

📄 PDF Abstract BibTeX arXiv:2205.01529

Code (3)

yzd-v/MGD 공식 구현 pytorch
PaddlePaddle/PaddleDetection paddle
yzd-v/cls_KD pytorch

Tasks

image-classificationImage ClassificationInstance SegmentationKnowledge Distillationobject-detectionObject DetectionSegmentationSemantic Segmentation

Methods 이 논문이 사용한 방법론

FPN 설명 없음
Spatial Pyramid Pooling Spatial Pyramid Pooling (SPP) is a pooling layer that removes the fixed-size constraint of the network, i.e. a CNN does not require a fixed-size input image. Specifically, we…
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
Focal Loss A Focal Loss function addresses class imbalance during training in tasks like object detection. Focal loss applies a modulating term to the cross entropy loss in order to…
Batch Normalization 설명 없음
RetinaNet RetinaNet is a one-stage object detection model that utilizes a focal loss function to address class imbalance during training.…
1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…
ASPP Atrous Spatial Pyramid Pooling (ASPP) is a semantic segmentation module for resampling a given feature layer at multiple rates prior to…

Similar Papers 제목 키워드 기반

Heterogeneous Generative Knowledge Distillation with Masked Image Modeling

2023-09-18 · ZiMing Wang, Shumin Han, Xiaodi Wang, Jing Hao 외

Small CNN-based models usually require transferring knowledge from a large model before they are deployed in computationally resource-limited edge devices. Masked image modeling (MIM) methods achieve great success in var…

image-classificationImage ClassificationKnowledge Distillationobject-detection+2

Masked Auto-Regressive Variational Acceleration: Fast Inference Makes Practical Reinforcement Learning

2025-11-19 · Yuxuan Gu, Weimin Bai, Yifei Wang, Weijian Luo 외 arxiv

Masked auto-regressive diffusion models (MAR) benefit from the expressive modeling ability of diffusion models and the flexibility of masked auto-regressive ordering. However, vanilla MAR suffers from slow inference due …

Reinforcement Learning

Di$\mathtt{[M]}$O: Distilling Masked Diffusion Models into One-step Generator

2025-03-19 · Yuanzhi Zhu, Xi Wang, Stéphane Lathuilière, Vicky Kalogeiton

Masked Diffusion Models (MDMs) have emerged as a powerful generative modeling technique. Despite their remarkable results, they typically suffer from slow inference with several steps. In this paper, we propose Di$\matht…

Conditional Image GenerationImage GenerationText to Image GenerationText-to-Image Generation

Self-distillation Augmented Masked Autoencoders for Histopathological Image Classification

2022-03-31 · Yang Luo, Zhineng Chen, Shengtian Zhou, Xieping Gao

Self-supervised learning (SSL) has drawn increasing attention in histopathological image analysis in recent years. Compared to contrastive learning which is troubled with the false negative problem, i.e., semantically si…

Cell SegmentationClassificationContrastive LearningHistopathological Image Classification+5

Partition Generative Modeling: Masked Modeling Without Masks

2025-05-24 · Justin Deschenaux, Lan Tran, Caglar Gulcehre

We introduce ``Partition Generative Models'' (PGMs), a novel approach to masked generative modeling (MGMs), particularly effective for masked diffusion language modeling (MDLMs). PGM divides tokens into two distinct grou…

Computational EfficiencyLanguage ModelingLanguage Modelling