Imagen de Portada

Gaussian Embeddings: How JEPAs Secretly Learn Your Data Density

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Balestriero, Randall, Ballas, Nicolas, Rabbat, Mike, LeCun, Yann
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Learning by Reconstruction Produces Uninformative Features For Perception
por: Balestriero, Randall, et al.
Publicado: (2024)

LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
por: Balestriero, Randall, et al.
Publicado: (2025)

Learning Latent Action World Models In The Wild
por: Garrido, Quentin, et al.
Publicado: (2026)

Revisiting Feature Prediction for Learning Visual Representations from Video
por: Bardes, Adrien, et al.
Publicado: (2024)

Learning and Leveraging World Models in Visual Representation Learning
por: Garrido, Quentin, et al.
Publicado: (2024)

A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures
por: Terver, Basile, et al.
Publicado: (2026)

Fast and Exact Enumeration of Deep Networks Partitions Regions
por: Balestriero, Randall, et al.
Publicado: (2024)

Intuitive physics understanding emerges from self-supervised pretraining on natural videos
por: Garrido, Quentin, et al.
Publicado: (2025)

Video Representation Learning with Joint-Embedding Predictive Architectures
por: Drozdov, Katrina, et al.
Publicado: (2024)

Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations
por: Kuang, Yilun, et al.
Publicado: (2026)

Blockwise Self-Supervised Learning at Scale
por: Siddiqui, Shoaib Ahmed, et al.
Publicado: (2023)

Stochastic positional embeddings improve masked image modeling
por: Bar, Amir, et al.
Publicado: (2023)

Variance-Covariance Regularization Improves Representation Learning
por: Zhu, Jiachen, et al.
Publicado: (2023)

LLM-JEPA: Large Language Models Meet Joint Embedding Predictive Architectures
por: Huang, Hai, et al.
Publicado: (2025)

Self-Supervised Anomaly Detection in the Wild: Favor Joint Embeddings Methods
por: Otero, Daniel, et al.
Publicado: (2024)

LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
por: Maes, Lucas, et al.
Publicado: (2026)

Navigation World Models
por: Bar, Amir, et al.
Publicado: (2024)

V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning
por: Mur-Labadia, Lorenzo, et al.
Publicado: (2026)

Transformers without Normalization
por: Zhu, Jiachen, et al.
Publicado: (2025)

On the Geometry of Deep Learning
por: Balestriero, Randall, et al.
Publicado: (2024)

Joint Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self Supervised Learning
por: Van Assel, Hugues, et al.
Publicado: (2025)

Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum
por: Lu, Wenquan, et al.
Publicado: (2025)

Interpreting Physics in Video World Models
por: Joseph, Sonia, et al.
Publicado: (2026)

$\mathbb{X}$-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs
por: Sobal, Vlad, et al.
Publicado: (2024)

World Models for Learning Dexterous Hand-Object Interactions from Human Videos
por: Goswami, Raktim Gautam, et al.
Publicado: (2025)

Whole-Body Conditioned Egocentric Video Prediction
por: Bai, Yutong, et al.
Publicado: (2025)

GPS-SSL: Guided Positive Sampling to Inject Prior Into Self-Supervised Learning
por: Feizi, Aarash, et al.
Publicado: (2024)

Deep Networks Always Grok and Here is Why
por: Humayun, Ahmed Imtiaz, et al.
Publicado: (2024)

FastDINOv2: Frequency Based Curriculum Learning Improves Robustness and Training Speed
por: Zhang, Jiaqi, et al.
Publicado: (2025)

A hierarchical loss and its problems when classifying non-hierarchically
por: Wu, Cinna, et al.
Publicado: (2017)

URLOST: Unsupervised Representation Learning without Stationarity or Topology
por: Yun, Zeyu, et al.
Publicado: (2023)

Scaling Language-Free Visual Representation Learning
por: Fan, David, et al.
Publicado: (2025)

Semantic Tube Prediction: Beating LLM Data Efficiency with JEPA
por: Huang, Hai, et al.
Publicado: (2026)

Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning
por: Zhai, Yuexiang, et al.
Publicado: (2024)

Causal-JEPA: Learning World Models through Object-Level Latent Masking
por: Nam, Heejeong, et al.
Publicado: (2026)

Your VAR Model is Secretly an Efficient and Explainable Generative Classifier
por: Chen, Yi-Chung, et al.
Publicado: (2025)

Understanding Contrastive Representation Learning from Positive Unlabeled (PU) Data
por: Acharya, Anish, et al.
Publicado: (2024)

UniBench: Visual Reasoning Requires Rethinking Vision-Language Beyond Scaling
por: Al-Tahan, Haider, et al.
Publicado: (2024)

Forgotten Polygons: Multimodal Large Language Models are Shape-Blind
por: Rudman, William, et al.
Publicado: (2025)

V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
por: Assran, Mido, et al.
Publicado: (2025)