| title | D-JEPA: A Decision-Aligned Latent World Model |
|---|---|
| authors | Shuaijun Liu, Chengyu Wu, Qifu Wen, Feiyang You, Chenglong Zhang, Shuyang Hao, Xi Lin, Ningxin Su |
| date | 2026-09-21 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2609.24749 |
| affiliation | NEBULIS Lab |
| upvotes | 16 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| primer | World models |
| abstract | Latent world models predict the consequences of actions, but accurate prediction does not guarantee that latent distance reflects which candidate will execute successfully. We identify a decision-local prediction gap: among the few futures competing for execution, a candidate predicted closer to the goal can produce a worse realized outcome than an available alternative. We introduce D-JEPA, a decision-aligned latent world model that learns decision-relevant relations among candidate futures from executed outcomes. A bounded, permutation-equivariant operator jointly reasons over goal-relative predictive features and ordinal evidence, refining pretrained predictive geometry where action choices are most consequential. Restricted predictor adaptation and a shared ordinal interface extend this alignment across complementary predictive geometries. D-JEPA further realizes the learned decision structure in JEPA-compatible future representations, enabling deployment through native latent-distance planning. Evaluations across latent control, manipulation, pretrained action-producing models, physical robots and autonomous driving demonstrate improved action selection, including 87.89% success on PushT, a 15.04-point average gain on RoboTwin, and a 17-point gain on physical robot tasks. These results establish decision-relevant relational structure as a direct bridge between predictive world modeling and effective control. |
cite
@misc{liu2026d,
title={D-JEPA: A Decision-Aligned Latent World Model},
author={Shuaijun Liu and Chengyu Wu and Qifu Wen and Feiyang You and Chenglong Zhang and Shuyang Hao and Xi Lin and Ningxin Su},
year={2026},
eprint={2609.24749},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2609.24749}
}