Tagged articles

representation collapse

4 articles · Page 1 of 1

Machine Learning Algorithms & Natural Language Processing

Jul 28, 2026 · Artificial Intelligence

How VISReg Overcomes JEPA’s Representation Collapse – LeCun’s Endorsement

VISReg introduces separate scale and shape regularizations based on sliced Wasserstein distance to prevent representation collapse in JEPA world models, achieving state‑of‑the‑art results on 15 benchmarks without heuristic tricks and matching DINOv2 performance with only one‑tenth of the data.

JEPAVISRegcomputer vision

0 likes · 16 min read

How VISReg Overcomes JEPA’s Representation Collapse – LeCun’s Endorsement

Machine Heart

Jul 14, 2026 · Artificial Intelligence

Why VISReg Excels: Solving Representation Collapse in JEPA Models

VISReg decouples regularization into independent scale and shape objectives using sliced Wasserstein distance, avoids gradient collapse, achieves state‑of‑the‑art results on 15 vision datasets with only about one‑tenth of the data, and offers linear computational complexity and strong OOD generalization.

JEPAVISRegrepresentation collapse

0 likes · 15 min read

Why VISReg Excels: Solving Representation Collapse in JEPA Models

Code Mala Tang

Apr 22, 2026 · Artificial Intelligence

How LeWorldModel Achieves Stable End‑to‑End World Modeling with Just Two Losses

LeWorldModel, a 2026 JEPA‑based world model introduced by Yann LeCun and collaborators, solves representation collapse with a minimalist two‑loss objective, delivering a 15‑million‑parameter system that trains in hours, runs 48× faster than prior baselines, and reaches near‑SOTA performance on robot control benchmarks.

Deep LearningJEPAWorld Model

0 likes · 6 min read

How LeWorldModel Achieves Stable End‑to‑End World Modeling with Just Two Losses

HyperAI Super Neural

Sep 15, 2025 · Artificial Intelligence

scSiameseClu Sets New SOTA on Unsupervised Single‑Cell Clustering Across 7 Datasets

The paper introduces scSiameseClu, a Siamese clustering framework that combines dual augmentation, siamese fusion, and optimal‑transport clustering to overcome representation collapse in scRNA‑seq data, and demonstrates state‑of‑the‑art performance on seven diverse single‑cell datasets and downstream annotation tasks.

ClusteringGraph Neural NetworkSiamese Network

0 likes · 11 min read

scSiameseClu Sets New SOTA on Unsupervised Single‑Cell Clustering Across 7 Datasets