mlsanityData to 14 Oct 2026

AdaJEPA: An Adaptive Latent World Model

arXiv 2606.32026 ↗First posted 30 Jun 2026cs.LG cs.AIcited by 20

On the map: World models → Latent predictive world models

Abstract (arXiv)

Latent world models enable planning from high-dimensional observations by predicting future states in a compact latent space. However, these models are typically kept frozen at test time: when their predictions become inaccurate, planning can fail, especially under test-time distribution shift. To address this, we propose AdaJEPA, an adaptive latent world model that performs test-time adaptation within the closed loop of model predictive control (MPC). After training, AdaJEPA plans and executes the first action chunk, uses the observed next-state transition as a self-supervised adaptation signal, and replans with the updated model. This closed-loop update continuously recalibrates the world model without additional expert demonstrations. Across a range of goal-reaching tasks, AdaJEPA substantially improves planning success with as few as one gradient step per MPC replanning step.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). Placed in Latent predictive world models, one of 11 directions in World models. Builds on 28 of its 45 references that are in the map. First posted on arXiv 30 Jun 2026.

Builds on

28 of 45 references.

Closest ancestors · 7

Same direction, newest first

  1. 2026Hierarchical Planning with Latent World Models3 Apr
  2. 2026LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels13 Mar
  3. 2026Temporal Straightening for Latent Planning12 Mar
  4. 2026Learning Invariant Visual Representations for Planning with Joint-Embedding Predictive World Models20 Feb
  5. 2025Closing the Train-Test Gap in World Models for Gradient-Based Planning10 Dec
  6. 2024DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning7 Nov
  7. 2022Temporal Difference Learning for Model Predictive Control9 Mar

Other references · 21

  1. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  2. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  3. 2020D4RL: Datasets for Deep Data-Driven Reinforcement Learningcited by 1,427
  4. 2021VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learningcited by 1,057
  5. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 809
  6. 2022Continual Test-Time Domain Adaptationcited by 525
  7. 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 512
  8. 2022Efficient Test-Time Model Adaptation without Forgettingcited by 469
  9. 2023Towards Stable Test-Time Adaptation in Dynamic Wild Worldcited by 415
  10. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 377
  11. 2025LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristicscited by 254
  12. 2022Test-Time Training with Masked Autoencoderscited by 212
9 more
  1. 2025AdaWorld: Learning Adaptable World Models with Latent Actionscited by 114
  2. 2024The Surprising Effectiveness of Test-Time Training for Few-Shot Learningcited by 96
  3. 2025Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Modelscited by 89
  4. 2024C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersioncited by 83
  5. 2023Test-Time Training on Video Streamscited by 37
  6. 2025AdaWM: Adaptive World Model based Planning for Autonomous Drivingcited by 24
  7. 2026Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMscited by 11
  8. 2025Adapting World Models with Latent-State Dynamics Residualscited by 9
  9. 2025Self-Improving Loops for Visual Robotic Planningcited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 20.

20 papers.

Sort:

By area:

  1. 2026Temporal-Distance JEPA: Plan-Aware Representation Learning for Latent World Model Predictive Controlcited by 8
  2. 2026LpWM: A Case for Sparse Representations in World Modelscited by 8
  3. 2026DWM: Separating World Effects from Actions in Latent World Modelscited by 3
  4. 2026Hydra-0: Action Flow for Generalist World Modeling and Controlcited by 3
  5. 2026A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inferencecited by 3
  6. 2026Sandwich-Residuals: Parameter-Efficient Test-time Adaptation of World Modelscited by 1
  7. 2026LaGSplat: Inferring Physics-Governed Interactive Simulation from Monocular Video Using Latent Lagrangian Gaussian Splattingcited by 0
  8. 2026Spectral-Target Physical Latent Structuring for JEPA-Style World Modelscited by 0
  9. 2026ARC-Bench: Closed-Loop Replanning Masks Broken Action Ranking in Frozen JEPA World Modelscited by 0
  10. 2026Underwater C3-JEPA: An Object-Centric Cross-View World Model for ROV Salvagecited by 0
  11. 2026AD-E2E-JEPA: A Joint-Embedding Predictive Architecture For End-to-End Autonomous Drivingcited by 0
  12. 2026Control-Geometry Straightening for Sampling-Based Latent Planningcited by 0
8 more
  1. 2026Beyond a single latent space: a dual-latent world model for long-horizon planningcited by 0
  2. 2026Do LLM Agents Execute the Plans They Declare? From Planning-Mode Declaration to Pattern-Specific Executioncited by 0
  3. 2026OverForge: Reasoning Through Strategies and Tactics Helps Cooperative Lifelong Adaptationcited by 0
  4. 2026JEPA-TTT: Persistent Test-Time Training of Latent World Models for Planning under Dynamics Shiftscited by 0
  5. 2026How Long, Not How Close: A Learned Temporal Metric for Planning in Latent World Modelscited by 0
  6. 2026H-JEPA: End-to-End Learning of Hierarchical World Models for Visual Planningcited by 0
  7. 2026Juno: Taming Predictive Latents for Vision-Language-Action Modelscited by 0
  8. 2026EvoBelief: Falsifiable World Models for Self-Improving LLM Agentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026SAGE: Subgoal-Conditioned Action Generation for Latent World Model Planning
  2. 2026Beyond the Next Step: Variable-Length Latent World Models for Long-Horizon Planning
  3. 2026Predictive but Not Plannable: RC-aux for Latent World Models
  4. 2026Enhancing Latent World Models with Associative Memory
  5. 2026Latent Geometry Beyond Search: Amortizing Planning in World Models
  6. 2026AdaReP:Adaptive Re-Planning under Model Mismatch for Neural World-Model Predictive Control
  7. 2026One-Step Prediction Fidelity Is a Poor Proxy for Planning Performance in JEPA World Models
  8. 2026DeepJEPA: Scaling World Models from Within