Generative Distribution Distillation

• URL: http://arxiv.org/abs/2507.14503v1
• Date: Sat, 19 Jul 2025 06:27:42 GMT
• Title: Generative Distribution Distillation
• Authors: Jiequan Cui, Beier Zhu, Qingshan Xu, Xiaogang Xu, Pengguang Chen, Xiaojuan Qi, Bei Yu, Hanwang Zhang, Richang Hong,
• Abstract summary: A naive textitGenDD baseline encounters two major challenges: the curse of high-dimensional optimization and the lack of semantic supervision from labels.<n>We introduce a textitSplit Tokenization strategy, achieving stable and effective unsupervised KD.<n>We also develop the textitDistribution Contraction technique to integrate label supervision into the reconstruction objective.
• Score: 109.84779053553896
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: In this paper, we formulate the knowledge distillation (KD) as a conditional generative problem and propose the \textit{Generative Distribution Distillation (GenDD)} framework. A naive \textit{GenDD} baseline encounters two major challenges: the curse of high-dimensional optimization and the lack of semantic supervision from labels. To address these issues, we introduce a \textit{Split Tokenization} strategy, achieving stable and effective unsupervised KD. Additionally, we develop the \textit{Distribution Contraction} technique to integrate label supervision into the reconstruction objective. Our theoretical proof demonstrates that \textit{GenDD} with \textit{Distribution Contraction} serves as a gradient-level surrogate for multi-task learning, realizing efficient supervised training without explicit classification loss on multi-step sampling image representations. To evaluate the effectiveness of our method, we conduct experiments on balanced, imbalanced, and unlabeled data. Experimental results show that \textit{GenDD} performs competitively in the unsupervised setting, significantly surpassing KL baseline by \textbf{16.29\%} on ImageNet validation set. With label supervision, our ResNet-50 achieves \textbf{82.28\%} top-1 accuracy on ImageNet in 600 epochs training, establishing a new state-of-the-art.

Related papers

• Graph-Based Uncertainty-Aware Self-Training with Stochastic Node Labeling [2.600103729157093]
  We propose a novel emphgraph-based uncertainty-aware self-training (GUST) framework to combat over-confidence in node classification.<n>Our method largely diverges from previous self-training approaches by focusing on emphstochastic node labeling grounded in the graph topology.<n> Experimental results on several benchmark graph datasets demonstrate that our GUST framework achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-03-26T21:54:19Z)
• Affinity-Graph-Guided Contractive Learning for Pretext-Free Medical Image Segmentation with Minimal Annotation [55.325956390997]
  This paper proposes an affinity-graph-guided semi-supervised contrastive learning framework (Semi-AGCL) for medical image segmentation. The framework first designs an average-patch-entropy-driven inter-patch sampling method, which can provide a robust initial feature space. With merely 10% of the complete annotation set, our model approaches the accuracy of the fully annotated baseline, manifesting a marginal deviation of only 2.52%.
  arXiv  Detail & Related papers  (2024-10-14T10:44:47Z)
• 2D Feature Distillation for Weakly- and Semi-Supervised 3D Semantic Segmentation [92.17700318483745]
  We propose an image-guidance network (IGNet) which builds upon the idea of distilling high level feature information from a domain adapted synthetically trained 2D semantic segmentation network. IGNet achieves state-of-the-art results for weakly-supervised LiDAR semantic segmentation on ScribbleKITTI, boasting up to 98% relative performance to fully supervised training with only 8% labeled points.
  arXiv  Detail & Related papers  (2023-11-27T07:57:29Z)
• Revisiting Class Imbalance for End-to-end Semi-Supervised Object Detection [1.6249267147413524]
  Semi-supervised object detection (SSOD) has made significant progress with the development of pseudo-label-based end-to-end methods. Many methods face challenges due to class imbalance, which hinders the effectiveness of the pseudo-label generator. In this paper, we examine the root causes of low-quality pseudo-labels and present novel learning mechanisms to improve the label generation quality.
  arXiv  Detail & Related papers  (2023-06-04T06:01:53Z)
• Neighbour Consistency Guided Pseudo-Label Refinement for Unsupervised Person Re-Identification [80.98291772215154]
  Unsupervised person re-identification (ReID) aims at learning discriminative identity features for person retrieval without any annotations. Recent advances accomplish this task by leveraging clustering-based pseudo labels. We propose a Neighbour Consistency guided Pseudo Label Refinement framework.
  arXiv  Detail & Related papers  (2022-11-30T09:39:57Z)
• One Class One Click: Quasi Scene-level Weakly Supervised Point Cloud Semantic Segmentation with Active Learning [29.493759008637532]
  We introduce One Class One Click (OCOC), a low cost yet informative quasi scene-level label, which encapsulates point-level and scene-level annotations. An active weakly supervised framework is proposed to leverage scarce labels by involving weak supervision from global and local perspectives. It considerably outperforms genuine scene-level weakly supervised methods by up to 25% in terms of average F1 score.
  arXiv  Detail & Related papers  (2022-11-23T01:23:26Z)
• STEdge: Self-training Edge Detection with Multi-layer Teaching and Regularization [15.579360385857129]
  We study the problem of self-training edge detection, leveraging the untapped wealth of large-scale unlabeled image datasets. We design a self-supervised framework with multi-layer regularization and self-teaching. Our method attains 4.8% improvement for ODS and 5.8% for OIS when tested on the unseen BIPED dataset.
  arXiv  Detail & Related papers  (2022-01-13T18:26:36Z)
• All-Around Real Label Supervision: Cyclic Prototype Consistency Learning for Semi-supervised Medical Image Segmentation [41.157552535752224]
  Semi-supervised learning has substantially advanced medical image segmentation since it alleviates the heavy burden of acquiring the costly expert-examined annotations. We propose a novel cyclic prototype consistency learning (CPCL) framework, which is constructed by a labeled-to-unlabeled (L2U) forward process and an unlabeled-to-labeled (U2L) backward process. Our framework turns previous textit"unsupervised" consistency into new textit"supervised" consistency, obtaining the textit"all-around real label supervision" property of our method.
  arXiv  Detail & Related papers  (2021-09-28T14:34:06Z)
• Mixed-supervised segmentation: Confidence maximization helps knowledge distillation [24.892332859630518]
  In this work, we propose a dual-branch architecture for deep neural networks. The upper branch (teacher) receives strong annotations, while the bottom one (student) is driven by limited supervision and guided by the upper branch. We show that the synergy between the entropy and KL divergence yields substantial improvements in performance.
  arXiv  Detail & Related papers  (2021-09-21T20:06:13Z)
• WSSOD: A New Pipeline for Weakly- and Semi-Supervised Object Detection [75.80075054706079]
  We propose a weakly- and semi-supervised object detection framework (WSSOD) An agent detector is first trained on a joint dataset and then used to predict pseudo bounding boxes on weakly-annotated images. The proposed framework demonstrates remarkable performance on PASCAL-VOC and MSCOCO benchmark, achieving a high performance comparable to those obtained in fully-supervised settings.
  arXiv  Detail & Related papers  (2021-05-21T11:58:50Z)
• Towards Unsupervised Sketch-based Image Retrieval [126.77787336692802]
  We introduce a novel framework that simultaneously performs unsupervised representation learning and sketch-photo domain alignment. Our framework achieves excellent performance in the new unsupervised setting, and performs comparably or better than state-of-the-art in the zero-shot setting.
  arXiv  Detail & Related papers  (2021-05-18T02:38:22Z)
• STRUDEL: Self-Training with Uncertainty Dependent Label Refinement across Domains [4.812718493682454]
  We propose an unsupervised domain adaptation (UDA) approach for white matter hyperintensity (WMH) segmentation. We propose to predict the uncertainty of pseudo labels and integrate it in the training process with an uncertainty-guided loss function to highlight labels with high certainty. Our results on WMH segmentation across datasets demonstrate the significant improvement of STRUDEL with respect to standard self-training.
  arXiv  Detail & Related papers  (2021-04-23T13:46:26Z)
• Towards Unbiased COVID-19 Lesion Localisation and Segmentation via Weakly Supervised Learning [66.36706284671291]
  We propose a data-driven framework supervised by only image-level labels to support unbiased lesion localisation. The framework can explicitly separate potential lesions from original images, with the help of a generative adversarial network and a lesion-specific decoder.
  arXiv  Detail & Related papers  (2021-03-01T06:05:49Z)
• MatchGAN: A Self-Supervised Semi-Supervised Conditional Generative Adversarial Network [51.84251358009803]
  We present a novel self-supervised learning approach for conditional generative adversarial networks (GANs) under a semi-supervised setting. We perform augmentation by randomly sampling sensible labels from the label space of the few labelled examples available. Our method surpasses the baseline with only 20% of the labelled examples used to train the baseline.
  arXiv  Detail & Related papers  (2020-06-11T17:14:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.