pith. sign in

arxiv: 2602.11389 · v2 · pith:IGW7QBTSnew · submitted 2026-02-11 · 💻 cs.AI

Causal-JEPA: Learning World Models through Object-Level Latent Masking

classification 💻 cs.AI
keywords c-jepamaskingobject-levelpredictionworldmodelsobject-centriccontrol
0
0 comments X
read the original abstract

World models require robust relational understanding to support prediction, reasoning, and control. While object-centric representations provide a useful abstraction, they are not sufficient to capture interaction-dependent dynamics. We therefore propose C-JEPA, a simple and flexible object-centric world model that extends masked joint embedding prediction from image patches to object-centric representations. By masking object-level latents and requiring each masked object state to be inferred from the surrounding context, C-JEPA imposes structured partial observability during training, creating counterfactual-like prediction queries that discourage shortcut solutions and make interaction-dependent prediction necessary under the learning objective. Empirically, C-JEPA leads to consistent gains in visual question answering, with an absolute improvement of about 20% in counterfactual reasoning over the same architecture without object-level masking. On agent control tasks, C-JEPA enables substantially more efficient planning by using only 1% of the total latent input features required by patch-based world models, while achieving comparable performance. Finally, we provide a formal analysis demonstrating that object-level masking induces useful inductive bias by controlling observability. Our code is available at https://github.com/galilai-group/cjepa.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. When Does LeJEPA Learn a World Model?

    stat.ML 2026-05 unverdicted novelty 8.0

    LeJEPA achieves linear identifiability of latent variables uniquely when the latents are Gaussian in worlds with stationary additive-noise transitions.

  2. Beyond the Next Step: Variable-Length Latent World Models for Long-Horizon Planning

    cs.LG 2026-06 unverdicted novelty 7.0

    VLWMs learn variable-length action-conditioned dynamics in latent space with curriculum training, yielding 13% average gains over prior latent world models on long-horizon tasks.

  3. Contrast encodes inductive bias: separating slow noise from dynamics in predictive representation learning

    cs.LG 2026-06 conditional novelty 7.0

    Cross-trajectory negative sampling in contrastive predictive objectives causes encoding of slow noise over dynamics; intra-trajectory sampling eliminates the shortcut and recovers dynamical variables even under strong noise.

  4. Latent State Design for World Models under Sufficiency Constraints

    cs.AI 2026-05 unverdicted novelty 7.0

    World models succeed when their latent states are built to meet task-specific sufficiency constraints rather than preserving the maximum amount of information.

  5. Slots, Transitions, Loops: Learning Composable World Models for ARC

    cs.CV 2026-06 unverdicted novelty 6.0

    Loop-OWM uses color-prototype slots, demonstration-conditioned task summaries, and looped transitions to model ARC rules as visual-symbolic state changes and outperforms baselines on ARC-1 and ARC-2.

  6. Trivium: Temporal Regret as a First-Class Objective for Causal-Memory Controllers

    cs.AI 2026-06 unverdicted novelty 6.0

    Trivium frames temporal regret as a first-class objective for causal-memory controllers and derives O(log E) bounds on miscalibration persistence under causal-probing assumptions, with preliminary validation on Causal...

  7. Slot-MPC: Goal-Conditioned Model Predictive Control with Object-Centric Representations

    cs.LG 2026-05 unverdicted novelty 6.0

    Slot-MPC learns slot representations to build a differentiable object-centric dynamics model that supports efficient gradient-based MPC for robotic manipulation in novel situations.

  8. LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

    cs.LG 2026-03 unverdicted novelty 6.0

    LeWM is the first end-to-end trainable JEPA from pixels that uses only two loss terms for stable training and fast planning on 2D/3D control tasks.

  9. Einstein World Models

    cs.AI 2026-06 unverdicted novelty 5.0

    Einstein World Models integrate visual rollouts from a callable world-module into LLM reasoning traces to support complex thought beyond language.

  10. Learning Sparse Latent Predictive Foundation Model for Multimodal Neuroimaging

    cs.CV 2026-06 unverdicted novelty 5.0

    Neuro-JEPA is a sparse multimodal foundation model pretrained on 1,551,862 brain MRI scans that shows stronger and more consistent performance than existing models and CNN baselines across 47 tasks from clinical and p...

  11. CausalVAE as a Plug-in for World Models: Towards Reliable Counterfactual Dynamics

    cs.LG 2026-04 unverdicted novelty 5.0

    CausalVAE plug-in for world models preserves factual prediction and boosts counterfactual retrieval, with large gains on physics benchmarks and recovered physical interaction trends.