Time-to-Goal: Learned Temporal Metrics for Long-Horizon Planning with JEPA World Models
cs.RO, cs.CV
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/yrluestc/T2G-JEPA
Terminology
Sources
- Revisiting Feature Prediction for Learning Visual Representations from Video
- LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
- H-JEPA: End-to-End Learning of Hierarchical World Models for Visual Planning
- DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning
- V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
- Mastering Diverse Domains through World Models
- Hierarchical Planning with Latent World Models
- Representation Learning with Contrastive Predictive Coding
- Learning to Reach Goals via Iterated Supervised Learning
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation