JEPA-x: Cross-Predictive Physics Grounding for Forecastable Latent Dynamics

[2608.24044] JEPA-x: Cross-Predictive Physics Grounding for Forecastable Latent Dynamics Skip to main content Search Submit Donate Log in Computer Science > Machine Learning arXiv:2608.24044 (cs) [Submitted on 25 Aug 2026 ( v1 ), last revised 28 Aug 2026 (this version, v3)] Title: JEPA-x: Cross-Predictive Physics Grounding for Forecastable Latent Dynamics Authors: Kehan Wen , Ziming Li , Siyuan Luo , Fan Shi View PDF HTML (experimental) Abstract: Latent world models plan by predicting how candidate actions advance learned latent dynamics. In self-predictive models, however, the encoder and predictor are optimized jointly and can co-adapt to latent transitions that are easy to predict but weakly constrained by the physical evolution of the scene. We introduce the cross-predictive JEPA (JEPA-x), which grounds latent dynamics in privileged physical trajectories. JEPA-x treats visual observations and physical states as corresponding views of the same action-conditioned trajectory, advances both through a shared predictor, and matches each prediction to the future representations of both modalities. This encourages the action-conditioned predictor to learn a common transition rule across the two views. Privileged physical state is used only during training, leaving a visual-only model at deployment. Empirical results show that JEPA-x reduces the rollout drift of a newly fitted predictor from $0.361$ to $0.104$ and increases mean control success from $53.6\%$ to $78.2\%$ on a multi-task suite spanning six evaluation subfamilies. We additionally show that direct physical-state regression improves decodability without improving forecastability or control, indicating that the benefit comes from shaping latent dynamics rather than merely encoding physical variables. Comments: Under review Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.24044 [cs.LG] (or arXiv:2608.24044v3 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.24044 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Kehan Wen [ view email ] [v1] Tue, 25 Aug 2026 04:07:07 UTC (10,487 KB) [v2] Wed, 26 Aug 2026 02:21:53 UTC (9,824 KB) [v3] Fri, 28 Aug 2026 10:39:50 UTC (9,813 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.LG < prev | next > new | recent | 2026-08 Change to browse by: cs References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) IArxiv recommender toggle IArxiv Recommender ( What is IArxiv? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)

Related documents

• JEPA-x: Cross-Predictive Physics Grounding for Forecastable Latent Dynamics#682838
• Validating Explainer Methods: A Functionally Grounded Approach for Numerical Forecasting#451792
• JEPA-TTT: Persistent Test-Time Training of Latent World Models for Planning under Dynamics Shifts#1155657
• FETS Benchmark: Foundation Models Enable Scalable and Generalizable Energy Time Series Forecasting#684431
• FETS Benchmark: Foundation Models Enable Scalable and Generalizable Energy Time Series Forecasting#682656