Self-Supervised Visual Representation Learning

Follow this topic Ask about this topic

Momentum

20 papers in the last four weeks, up 100% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 151

All topics

Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss
Oct 7, 2026Ruoyu Zhao, Yuting Chen, Jinheng Zhang +2Contrastive Learning Self-Supervised Visual Representation Learning
City University of Hong Kong · Georgia Institute of Technology · University of Pennsylvania +2
Scalable Patch-Level Self-Supervised Learning
Oct 7, 2026Maximilian Seitzer, Gabriele Trivigno, Antonín Vobecký +5Self-Supervised Learning Self-Supervised Visual Representation Learning
Meta FAIR
DisParQ: Self-Supervised Part Concepts for Interpretable Vision Foundation Models
Oct 7, 2026Adam Pardyl, Siddhartha Gairola, Sukrut Rao +4Explainable Image Classification Vision Foundation Models
Jagiellonian University, Faculty of Mathematics and Computer Science · Jagiellonian University, Doctoral School of Exact and Natural Sciences · Max Planck Institute for Informatics, Saarland Informatics Campus +2
Unlocking Fine-Grained Perception in CLIP via Structurally-Aware Latent Masked Modeling
Oct 6, 2026Juntong Li, Lingwei Dang, Haomin Wu +3VLM Adaptation Dense Prediction
School of Software Engineering, South China University of Technology
From Pixels, Without Pre-training: Joint Generative and Self-Supervised Representation Learning in One Model
Oct 5, 2026Vicente Balmaseda, Ching-Long Lin, Tianbao YangRepresentation Learning Image Generation
Department of Computer Science and Engineering Texas A&M University College Station, TX, USA · Mechanical Engineering University of Iowa Iowa City, IA, USA
CI-JEPA: A Counterfactual Analysis of Latent Representations in Joint-Embedding Predictive Architectures for Self-Supervised Learning
Oct 4, 2026Mintu Dutta, Ritesh Vyas, Mohendra Roy *Causal Representation Learning Joint-Embedding Predictive Architecture
Department of Information and Communication Technology, Pandit Deendayal Energy University, Gandhinagar, Gujarat, India · Department of Electrical and Electronics Engineering, Birla Institute of Technology & Science, Pilani, Rajasthan, India
The hidden advantage of mask resampling: a theory of masked autoencoders
Oct 1, 2026Jorge Medina Moreira, Lorenzo Bardone, Lenka ZdeborováMasked Autoencoders Masked Language Modeling
Statistical Physics of Computation Laboratory, École Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland
Increasing Width Allows Greedy Layer-wise Training to Rival End-to-End Backpropagation in Self-Supervised Learning
Sep 30, 2026Syon Mansur, Joel ZylberbergDeep Learning Optimization Convolutional Neural Networks
Jules Stein Eye Institute University of California, Los Angeles Los Angeles, CA 90095, USA
I Have a Stream: Making Self-Supervised Learning Work on Continuous Video
Sep 30, 2026Ivan Martinović, Lukas Knobel, Yuki M. AsanoMasked Autoencoders Self-Supervised Learning
Faculty of Electrical Engineering and Computing, University of Zagreb · Fundamental AI Lab, University of Technology Nuremberg
Emergent Multi-View Geometry Through Self-Distillation
Sep 30, 2026David Nordström, Thibaut Loiseau, Vincent Lepetit +3Multi-View Geometry Multi-View 3D Reconstruction
Chalmers University of Technology, Sweden · LIGM, Ecole des Ponts, Univ. Gustave Eiffel, CNRS, France · Linköping University, Sweden +1
Masked Swingers: Harnessing Data Augmentation to Advance Autoencoders for Self-Supervised Learning
Sep 29, 2026Anthony Fuller, Scott C. Lowe, Daniel G. Kyrollos +3Masked Autoencoders Invariant Representation Learning
Vector Institute · Carleton University · University of Guelph +1
End-to-End Self-Supervised RGB-T Tracking without Modality Misleading
Sep 29, 2026Shenglan Li, Rui Yao, Kunyang Sun +4Multimodal Robustness Visual Object Tracking
School of Computer Science and Technology / School of Artificial Intelligence, China University of Mining and Technology, China. · Mine Digitization Engineering Research Center of the Ministry of Education, China. · University of Auckland, Auckland, New Zealand. +1
Representation by Design in Generation: Cross-View Class-Token Alignment in Diffusion Transformers
Sep 28, 2026Xiaoyu Wu, Yifei Wang, Chen WeiFlow Matching Diffusion Transformer
Carnegie Mellon University · Rice University
Sparse-View Interpretable 3D Animal Behavior Representations for Neural Encoding and Decoding
Sep 28, 2026Xinming Dai, Qihang Jin, Tianshu Tan +8Representation Learning Sparse-View 3D Reconstruction
Columbia University · University of Science and Technology of China · Harvard University +3
W2Rep: Learning Visual Representations by Watching the World Change
Sep 28, 2026Wen Huang, Hang Guo, Jiarui Yang +3Video Representation Learning Self-Supervised Visual Representation Learning
Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China · Nankai University, Tianjin, China · Shenzhen University, Shenzhen, China
Generative Uncertainty as a Self-supervised Signal for Semantic Similarity Learning
Sep 28, 2026Enrico Pallotta, Sina Raoufi, Lars Doorenbos +2Video Representation Learning Feature Selection
University of Bonn · Lamarr Institute for ML & AI · ENSTA Paris, Institut Polytechnique de Paris
$λ$ -JEPA Spectral Anti-Collapse Regularization for Self-Supervised Learning
Sep 28, 2026Berker Demirel, Clémentine Dominé, Valentino Maiorca +3Spectral Regularization Joint-Embedding Predictive Architecture
Institute of Science and Technology Austria (ISTA) Am Campus 1, 3400 Klosterneuburg, Austria
A Light Bilevel Refinement Aligns Self-Supervised Representations for Stronger Task-Specific Learning
Sep 27, 2026Gustav Wagner Zakarias, Zheng-Hua TanFine-Tuning Bilevel Optimization
Aalborg University · Pioneer Centre for Artificial Intelligence, Denmark
When Does Geometric View Synthesis Help Wine Label Retrieval? A Public One-Shot Benchmark Across Self-Supervised and Vision-Language Backbones
Sep 27, 2026Yueh-Cheng HuangVLM Adaptation Novel View Synthesis
Department of Computer Science and Information Engineering, National Dong Hwa University, Hualien 974301, Taiwan
Conditional Predictive Sufficient Statistics for Visual Representation Learning
Sep 25, 2026Yuzhou HongVisual Representation Learning Predictive Representation Learning
Zhejiang Sci-Tech University
Learning a Flow to Self-Supervised Representations
Sep 24, 2026Yuling Jiao, Wensen Ma, Houduo Qi +1Flow Matching Representation Learning
School of Artificial Intelligence, National Center for Applied Mathematics in Hubei, Hubei Key Laboratory of Computational Science, Wuhan University, Wuhan, China. · Department of Applied Mathematics, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong SAR, China. · Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong SAR, China.
Two Global Crops Suffice: Locating Semantic Emergence in DINO-Style Self-Supervised Learning
Sep 23, 2026Basavaraj Sunagad, Artur Jesslen, Adam KortylewskiMulti-View Learning Self-Distillation
CISPA Helmholtz Center for Information Security · University of Freiburg
Positive Pair Geometry Matters: Optimal Transport for Contrastive Learning of Visual Representations
Sep 21, 2026Akshit Nanda, Shahzad Ahmad, Ram Prasad PadhyContrastive Learning Data Augmentation
Department of Computer Science and Engineering, Indian Institute of Technology Bhubaneswar · Østfold University of Applied Sciences, NTNU
Vision Transformers versus convolutional neural networks for fine-grained orchid genus identification in a species-rich, data-poor flora: a controlled benchmark on the Orchidaceae of New Guinea
Sep 21, 2026Reza Saputra, Diah Harnoni Apriyanti, André Schuiteman +4Vision Transformer Fine-Grained Image Retrieval
College of Science and Engineering, James Cook University. McGregor Rd, Smithfield, Cairns, QLD 4878, Australia · Southwest Papua Natural Resources Conservation Agency, Ministry of Forestry, Indonesia. Jalan Klamono KM 16, Sorong, Southwest Papua Province, Indonesia · The Directorate of Scientific Collection Management, National Research and Innovation Agency (BRIN), Republic of Indonesia, Gedung Kehati, KST Soekarno BRIN, Jl. Raya Jakarta - Bogor KM 46, Cibinong, Kabupaten Bogor, Jawa Barat, 16911, Indonesia. +5
ParticleSplat: Self-supervised Object-centric Latent Particle Splatting
Sep 16, 2026Lyuxing He, Daniel Guo, Elizabeth Terveen +33D Gaussian Splatting Object-Centric Representation Learning
Robotics Institute, Carnegie Mellon University
CoViT: Instance-Correspondence Contrastive Learning for Vision Transformer
Sep 1, 2026Yisen Wang, Zhirong Wu, Limin WangContrastive Learning Instance Segmentation
State Key Laboratory for Novel Software Technology Nanjing University, China
Benchmarking Spatial, Spectral, and Self-Supervised Cues for Face Forgery Detection under Realistic Degradation
Sep 1, 2026Lucas Cunha, Lucas Sotomaior, Lucas Gasperin +3Image Corruption Robustness Image Forgery Detection
Pontifical Catholic University of Paraná, Curitiba, Brazil
Pix2Rep-v2: Data-Efficient Representation Learning for Dense Medical Imaging Applications
Sep 1, 2026S. Sifaoui, E. Angelini, S. Toupin +2Representation Learning Visual Representation Learning
LTCI, Télécom Paris, Institut Polytechnique de Paris, Palaiseau, France · MIRACL.ai Laboratory, Hôpital Universitaire Lariboisière (AP-HP), Paris, France · Université Paris Cité, Inserm MASCOT, Paris, France
CMRVision: A Foundation Model for Cardiac MR Image Analysis
Sep 1, 2026Athira J. Jacob, Puneet Sharma, Daniel RueckertMedical Imaging Foundation Models Medical Image Classification
Digital Technology and Innovation, Siemens Healthineers, Princeton, NJ, USA · Chair for Al in Healthcare and Medicine, Technical University of Munich (TUM) and TUM University Hospital, Munich, Germany · Department of Computing, Imperial College London, UK
ViTAMINS: An Empirical Study of Training Self-Supervised Vision Transformers with Synthetic Hard Negatives
Sep 1, 2026Nikos Giakoumoglou, Andreas Floros, Kleanthis-Marios Papadopoulos +1Contrastive Learning Vision Transformer
Imperial College London

Oct 7, 2026, cs.LGJ/K move · Enter open · S save

Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss

Ruoyu Zhao, Yuting Chen, Jinheng Zhang, Zhehao Zou+1

City University of Hong Kong · Georgia Institute of Technology · University of Pennsylvania+2

Contrastive Learning Self-Supervised Visual Representation Learning Latent Space Alignment

Open paper AskPDF arXiv

What can a distribution-matching regularizer such as SIGReg in LeJEPA certify about contrastive learning? We study shared Gaussianization (SG), a characteristic-function Gaussianity test on the average of two normalized views, scaled by an independent

χ_d

radius. Because disagreeing views shorten the average, one test detects both misalignment and non-uniformity. SG vanishes exactly at the aligned, uniform minimizers of population InfoNCE, and under equal marginals it bounds the InfoNCE excess by

4\cdot 3^{3/4}β

times the square root of the SG loss, plus a term linear in the loss. The square-root rate and this dimension-free constant are sharp, and no squared mean-embedding distance on view pairs achieves a faster rate. With an explicit alignment term, a rotation-invariant uniformity test gives a linear bound if and only if its spectrum dominates that of InfoNCE's kernel

e^{βu^\top v}

; SG's own test does, Gaussian kernels

e^{-γ\|u-v\|^2}

qualify exactly when

γ\ge β/2

, and moment matching never does. Away from the optimum, the objectives differ. Along an isotropic nuisance channel, pure SG lowers its loss by adding per-view nuisance whenever the shared code is non-uniform. An alignment weight above the channel's gain makes the nuisance-free solution a strict local minimizer; for LeJEPA, the same rule gives a critical SIGReg weight that decreases with the batch size. At finite batch size, an off-diagonal U-statistic removes a plug-in bias toward misalignment. In controlled latent-variable models, pure SG retains per-view style, an alignment weight above the measured gain removes it, and for LeJEPA at three batch sizes the measured gain separates the encoders that retain style from those that do not. InfoNCE training also reaches a lower SG

_{0.2}

loss than SG

_{0.2}

training from scratch, which points to an optimization gap.

Momentum

Related topics

Latest papers 151