Dataset distillation compresses large datasets into compact synthetic ones to reduce storage and computational costs. Among various approaches, distribution matching (DM)-based methods have attracted attention for their high efficiency. However, they often overlook the evolution of feature representations during training, which limits the expressiveness of synthetic data and weakens downstream performance. To address this issue, we propose Trajectory Guided Dataset Distillation (TGDD), which reformulates distribution matching as a dynamic alignment process along the model's training trajectory. At each training stage, TGDD captures evolving semantics by aligning the feature distribution between the synthetic and original dataset. Meanwhile, it introduces a distribution constraint regularization to reduce class overlap. This design helps synthetic data preserve both semantic diversity and representativeness, improving performance in downstream tasks. Without additional optimization overhead, TGDD achieves a favorable balance between performance and efficiency. Experiments on ten datasets demonstrate that TGDD achieves state-of-the-art performance, notably a 5.0% accuracy gain on high-resolution benchmarks.
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Nanjing University of Science and Technology（南京理工大学）
Deep models trained on long-tailed datasets exhibit unsatisfactory performance on tail classes. Existing methods usually modify the classification loss to increase the learning focus on tail classes, which unexpectedly sacrifice the performance on head classes. In fact, this scheme leads to a contradiction between the two goals of long-tailed learning, i.e., learning generalizable representations and facilitating learning for tail classes. In this work, we explore knowledge distillation in long-tailed scenarios and propose a novel distillation framework, named Balanced Knowledge Distillation (…
Real-world imagery is often characterized by a significant imbalance of the number of images per class, leading to long-tailed distributions. An effective and simple approach to long-tailed visual recognition is to learn feature representations and a classifier separately, with instance and class-balanced sampling, respectively. In this work, we introduce a new framework, by making the key observation that a feature representation learned with instance sampling is far from optimal in a long-tailed setting. Our main contribution is a new training method, referred to as Class-Balanced Distillati…
VICO Group, University of Edinburgh（爱丁堡大学VICO研究组）；University of Edinburgh（爱丁堡大学）；United Kingdom（英国）
Multi-task learning (MTL) is to learn one single model that performs multiple tasks for achieving good performance on all tasks and lower cost on computation. Learning such a model requires to jointly optimize losses of a set of tasks with different difficulty levels, magnitudes, and characteristics (e.g. cross-entropy, Euclidean loss), leading to the imbalance problem in multi-task learning. To address the imbalance problem, we propose a knowledge distillation based method in this work. We first learn a task-specific model for each task. We then learn the multi-task model for minimizing task-…
Deep learning has achieved remarkable progress for visual recognition on large-scale balanced datasets but still performs poorly on real-world long-tailed data. Previous methods often adopt class re-balanced training strategies to effectively alleviate the imbalance issue, but might be a risk of over-fitting tail classes. The recent decoupling method overcomes over-fitting issues by using a multi-stage training scheme, yet, it is still incapable of capturing tail class information in the feature learning stage. In this paper, we show that soft label can serve as a powerful solution to incorpor…
Modality-Balanced Learning for Multimedia Recommendation
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）
Aiming towards human-level generalization, there is a need to explore adaptable representation learning methods with greater transferability. Most existing approaches independently address task-transferability and cross-domain adaptation, resulting in limited generalization. In this paper, we propose UM-Adapt - a unified framework to effectively perform unsupervised domain adaptation for spatially-structured prediction tasks, simultaneously maintaining a balanced performance across individual tasks in a multi-task setting. To realize this, we propose two novel regularization strategies; a) Con…
Learn from Balance: Rectifying Knowledge Transfer for Long-Tailed Scenarios
School of Computer Science and Technology, Southwest University of Science and Technology（西南科技大学计算机科学与技术学院）；School of Information Science and Technology, Great Bay University（大湾区大学信息科学与技术学院）；Graduate School of Science and Engineering, Hosei University（法政大学理工学研究院）
HAD: Heterogeneity-Aware Distillation for Lifelong Heterogeneous Learning
Southern University of Science and Technology（南方科技大学）；Zhejiang University（浙江大学）；Technical University of Munich（慕尼黑工业大学）；City University of Hong Kong（香港城市大学）；Shanghai Jiao Tong University（上海交通大学）
Southern University of Science and Technology（南方科技大学）；The Chinese University of Hong Kong（香港中文大学）；Xiamen University（厦门大学）；City University of Hong Kong（香港城市大学）
The amount of medical images for training deep classification models is typically very scarce, making these deep models prone to overfit the training data. Studies showed that knowledge distillation (KD), especially the mean-teacher framework which is more robust to perturbations, can help mitigate the over-fitting effect. However, directly transferring KD from computer vision to medical image classification yields inferior performance as medical images suffer from higher intra-class variance and class imbalance. To address these issues, we propose a novel Categorical Relation-preserving Contr…
National University of Defense Technology（国防科技大学）；University of Oulu（奥卢大学）
Given a model well-trained with a large-scale base dataset, Few-Shot Class-Incremental Learning (FSCIL) aims at incrementally learning novel classes from a few labeled samples by avoiding overfitting, without catastrophically forgetting all encountered classes previously. Currently, semi-supervised learning technique that harnesses freely-available unlabeled data to compensate for limited labeled data can boost the performance in numerous vision tasks, which heuristically can be applied to tackle issues in FSCIL, i.e., the Semi-supervised FSCIL (Semi-FSCIL). So far, very limited work focuses o…
Structural Assessment for Understanding and Guiding Dataset Distillation in Discrete Token Space
The University of British Columbia（不列颠哥伦比亚大学）；The Ohio State University（俄亥俄州立大学）；Czech Technical University in Prague（布拉格捷克理工大学）；TerraSense Analytics（TerraSense Analytics公司）
University of Toronto（多伦多大学）；Mila（米拉研究所）；Université de Montréal（蒙特利尔大学）
An effective framework for learning 3D representations for perception tasks is distilling rich self-supervised image features via contrastive learning. However, image-to point representation learning for autonomous driving datasets faces two main challenges: 1) the abundance of self-similarity, which results in the contrastive losses pushing away semantically similar point and image regions and thus disturbing the local semantic structure of the learned representations, and 2) severe class imbalance as pretraining gets dominated by over-represented classes. We propose to alleviate the self-sim…
University of Texas at Austin（德克萨斯大学奥斯汀分校）；University of Colorado Boulder（科罗拉多大学博尔德分校）
Photorealistic style transfer entails transferring the style of a reference image to another image so the result seems like a plausible photo. Our work is inspired by the observation that existing models are slow due to their large sizes. We introduce PCA-based knowledge distillation to distill lightweight models and show it is motivated by theory. To our knowledge, this is the first knowledge distillation method for photorealistic style transfer. Our experiments demonstrate its versatility for use with different backbone architectures, VGG and MobileNet, across six image resolutions. Compared…
Mainstream object detectors are commonly constituted of two sub-tasks, including classification and regression tasks, implemented by two parallel heads. This classic design paradigm inevitably leads to inconsistent spatial distributions between classification score and localization quality (IOU). Therefore, this paper alleviates this misalignment in the view of knowledge distillation. First, we observe that the massive teacher achieves a higher proportion of harmonious predictions than the lightweight student. Based on this intriguing observation, a novel Harmony Score (HS) is devised to estim…
Indian Institute of Science（印度科学学院）；Indian Institute of Technology（印度理工学院）
Knowledge Distillation is an effective method to transfer the learning across deep neural networks. Typically, the dataset originally used for training the Teacher model is chosen as the "Transfer Set" to conduct the knowledge transfer to the Student. However, this original training data may not always be freely available due to privacy or sensitivity concerns. In such scenarios, existing approaches either iteratively compose a synthetic set representative of the original training dataset, one sample at a time or learn a generative model to compose such a transfer set. However, both these appr…
Beyond Modality Collapse: Representations Blending for Multimodal Dataset Distillation
Centre for Frontier AI Research, Agency for Science, Technology and Research, Singapore（前沿人工智能研究中心，科技研究局，新加坡）；Institute of High Performance Computing, Agency for Science, Technology and Research, Singapore（高性能计算研究所，科技研究局，新加坡）；National University of Singapore, Singapore（新加坡国立大学）；Zhejiang University, China（浙江大学）
Privileged Lesion-Context Relational Distillation for Mask-Free Skin Lesion Classification
Victoria University（维多利亚大学）；Bangaldesh University of Business and Technology（孟加拉国商业与技术大学）；Daffodil Institute of Information Technology（达芙妮信息技术学院）；American International University-Bangladesh (AIUB)（孟加拉国美国国际大学）；Multimedia University（多媒体大学）
A Graph-Augmented knowledge Distillation based Dual-Stream Vision Transformer with Region-Aware Attention for Gastrointestinal Disease Classification with Explainable AI
Department of Computer Science and Engineering（计算机科学与工程系）
AlignUS: MRI-Guided Ultrasound Representation Learning for ALS Classification from Tongue Images
Unité de Recherche en Systèmes Intelligents Avancés (URSIA), Institut Supérieur du Numérique (SupNum)（高级智能系统研究单位（URSIA），数字高等学院（SupNum））；University of Basel（巴塞尔大学）
Incremental Learning (IL) allows AI systems to adapt to streamed data. Most existing algorithms make two strong hypotheses which reduce the realism of the incremental scenario: (1) new data are assumed to be readily annotated when streamed and (2) tests are run with balanced datasets while most real-life datasets are actually imbalanced. These hypotheses are discarded and the resulting challenges are tackled with a combination of active and imbalanced learning. We introduce sample acquisition functions which tackle imbalance and are compatible with IL constraints. We also consider IL as an imb…
Mind Your Margin and Boundary: Are Your Distilled Datasets Truly Robust?
The Laboratory of Intelligent Collaborative Computing of UESTC, Chengdu, China（UESTC智能协同计算实验室，中国成都）；Monash University, Melbourne, Australia（墨尔本大学，澳大利亚墨尔本）；Guangdong Laboratory of Artificial Intelligence（广东人工智能实验室）
Long-Tailed Recognition on Binary Networks by Calibrating A Pre-trained Model
Minds and Company（明兹公司）；Upstage AI（上阶段人工智能公司）；Gwangju Institute of Science and Technology (GIST)（光州科学技术院）；Alsemy（阿尔塞米公司）；Seoul National University（首尔大学）
Online continual learning in the wild is a very difficult task in machine learning. Non-stationarity in online continual learning potentially brings about catastrophic forgetting in neural networks. Specifically, online continual learning for autonomous driving with SODA10M dataset exhibits extra problems on extremely long-tailed distribution with continuous distribution shift. To address these problems, we propose multiple deep metric representation learning via both contrastive and supervised contrastive learning alongside soft labels distillation to improve model generalization. Moreover, w…
Peking University Shenzhen Graduate School（北京大学深圳研究生院）；King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）
Class-Incremental Learning (CIL) struggles with catastrophic forgetting when learning new knowledge, and Data-Free CIL (DFCIL) is even more challenging without access to the training data of previously learned classes. Though recent DFCIL works introduce techniques such as model inversion to synthesize data for previous classes, they fail to overcome forgetting due to the severe domain gap between the synthetic and real data. To address this issue, this paper proposes relation-guided representation learning (RRL) for DFCIL, dubbed R-DFCIL. In RRL, we introduce relational knowledge distillation…
In the real world, medical datasets often exhibit a long-tailed data distribution (i.e., a few classes occupy the majority of the data, while most classes have only a limited number of samples), which results in a challenging long-tailed learning scenario. Some recently published datasets in ophthalmology AI consist of more than 40 kinds of retinal diseases with complex abnormalities and variable morbidity. Nevertheless, more than 30 conditions are rarely seen in global patient cohorts. From a modeling perspective, most deep learning models trained on these datasets may lack the ability to gen…
VDOT++: Unified Few-Step Video Generation via Unbalanced Optimal Transport Distillation
University of Sydney（悉尼大学）；Shanghai AI Laboratory（上海人工智能实验室）；Hong Kong University of Science and Technology（香港科技大学）；Chinese University of Hong Kong（香港中文大学）
Privacy-Preserving Synthetic Continual Semantic Segmentation for Robotic Surgery
National University of Singapore（新加坡国立大学）；NUSRI Suzhou（新加坡国立大学苏州研究院）；University College London（伦敦大学学院）；The Chinese University of Hong Kong（香港中文大学）；Shun Hing Institute of Advanced Engineering（信兴高等工程研究院）
