mlsanityData to 14 Oct 2026

J-LAW: Joint Localization and Action-Conditioned World Modeling via Coupled Latent Factor Graphs

arXiv 2606.28712 ↗First posted 27 Jun 2026cs.RO cs.LGcited by 0

On the map: World models → Latent predictive world models

Abstract (arXiv)

Classical simultaneous localization and mapping (SLAM) estimates metric poses and a geometric map but does not provide an action-conditioned predictive state. Action-conditioned world models learn compact latent dynamics but ignore global metric consistency and accumulate drift under open-loop rollout. We introduce J-LAW (Joint Localization and Action-Conditioned World Modeling), a unified factor-graph formulation that connects metric pose variables, predictive latent states, and persistent latent landmarks in this letter.J-LAW represents each image as a compact predictive state and combines it with pose or motion measurements through a separately learned mapping. Its maximum a posteriori (MAP) factor graph enforces consistency between these complementary sources of information over time. Experiments on PushT and WildGS show that J-LAW's factor-graph representation can improve long-horizon latent consistency and recover more reliable predictive states under partial observations, forming a foundation for future integrated localization and planning systems.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Latent predictive world models, one of 11 directions in World models. Builds on 7 of its 11 references that are in the map. First posted on arXiv 27 Jun 2026.

Builds on

7 of 11 references.

Closest ancestors · 2

Same direction, newest first

  1. 2026SkyJEPA: Learning Long-Horizon World Models for Zero-Shot Sim-to-Real Control of Quadrotors22 Jun
  2. 2026LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels13 Mar

Other references · 5

  1. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  2. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 809
  3. 2026V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learningcited by 117
  4. 2025WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environmentscited by 50
  5. 2021LatentSLAM: unsupervised multi-sensor representation learning for localization and mappingcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026World SLAM Model: Joint World Modeling for SLAM and Navigation
  2. 2026DynoSLAM: Dynamic SLAM with Generative Graph Neural Networks for Real-World Social Navigation
  3. 2026Toward Physically Grounded JEPA World Models for Goal-Conditioned Robotic Planning
  4. 2026Does Latent Planning Survive Point Clouds? Action-Conditioned JEPA World Models for Geometric Observations
  5. 2026LaWM: Least Action World Models for Long-Horizon Physical Consistency from Visual Observations
  6. 2026Frozen in a Frame: The Velocity Blind Spot in JEPA World Models
  7. 2026G2WAM: World Action Models for Efficient and Geometry Grounded Image-Goal Navigation
  8. 2026LA-Pose: Latent Action Pretraining Meets Pose Estimation