arXiv Machine Learning
Preserving Unstable Modes Through Inverse Dynamics in JEPA World Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Jun 19
Sensorimotor World Models: Perception for Action via Inverse Dynamics
arXiv:2606. 20104v1 Announce Type: cross Abstract: Perception for action suggests that representations of the world should be shaped not by visual fidelity alone, but by their relevance for actions.
arXiv AI
Jul 1
Delta-JEPA: Learning Action-Sensitive World Models via Latent Difference Decoding
arXiv:2606. 31232v1 Announce Type: new Abstract: Learning visual world models for planning requires compact latent dynamics that remain sensitive to actions, yet reconstruction-free joint-embedding objectives can collapse to action-insensitive representations.
arXiv Computer Vision
Sep 22
MotionJEPA: Preventing Temporal Feature Collapse by Capturing Visual Changes in Latent Space
arXiv:2609.23881v1 Announce Type: new Abstract: Joint Embedding Predictive Architectures (JEPAs) are a promising paradigm for learning task-agnostic latent world models without visual reconstruction....
arXiv AI
Jul 14
Towards Predictive, Aligned, and Scalable Robot Learning
arXiv:2607. 11270v1 Announce Type: cross Abstract: Learning, at its core, extends beyond memorization to the ability to reason and solve novel problems by navigating a space of possibilities.
arXiv Computer Vision
Sep 30
WorldLine: Action-Driven Visual Simulation for Robotic Manipulation
arXiv:2609.38059v1 Announce Type: cross Abstract: Real-world robot learning is constrained by the cost of collecting experience and evaluating candidate behaviors. Video generation models offer a sca...
arXiv Machine Learning
Aug 7
Observation-Grounded Self-Predictive Reinforcement Learning for Visual Continuous Control
arXiv:2608. 05989v1 Announce Type: new Abstract: Sample-efficient policy learning from pixels is a long-standing challenge in reinforcement learning (RL).