AI Briefing#jepa
The latest AI and developer news about #jepa, with the original source and a short summary.
The latest AI and developer news about #jepa, with the original source and a short summary.
The self-supervised encoder achieves 49.3% tactile-to-video recall at 4K hours, significantly outperforming reconstruction-based baselines.
A new theorem demonstrates that Joint-Embedding Predictive Architectures (JEPAs) achieve cross-modal alignment with 42 times fewer training samples than next-token prediction.
Research demonstrating applications across various domains, including a 31.2% code repair success rate, notes that the 99% GUI success rate is based on policy ranking.
F.R.A.N.T.Z. extracts scene semantics from monocular video without synthetic data or task-specific training.
A technical guide that implements Yann LeCun's JEPA architecture from scratch while explaining its underlying principles.
AMI Labs, founded by Yann LeCun, is developing a new AI architecture called JEPA to overcome the limitations of LLMs.
By using action-prefix prediction instead of sequential prediction in JEPA-based world models, robot planning efficiency has been improved.
A small open source world model that uses the JEPA architecture to predict the next state of video has been released.
Sub-JEPA boosts LeWM performance by splitting the high-dimensional Gaussian constraint and applying it separately.
LeCun believes world models, not LLMs, are the key to real-world intelligence.