The paper introduces a diagnostic protocol to examine how passive object-state world models encode event-conditioned latent physical structure. It evaluates GRU, Transformer-lite, and RSSM-lite models on a dataset featuring free motion, collision, and occlusion events, finding that each architecture learns predictive dynamics and that event context shifts the emphasis among kinematic, contact, and object-permanence readouts. Functional sensitivity tests reveal contact-related structure during collisions and object-permanence structure during occlusions, supporting the presence of event-conditioned latent structure without explicit physical modules.
arXiv:2606. 28455v1 Announce Type: cross Abstract: World models can predict future physical states, but prediction accuracy alone does not explain how physical information is organized and used inside their latent dynamics.
arXiv:2608. 20009v1 Announce Type: new Abstract: Understanding object dynamics requires not only predicting future trajectories but also examining whether a model captures the physical properties that govern motion.
arXiv:2606. 13053v1 Announce Type: cross Abstract: Pretrained-feature world models provide a useful substrate for robot imagination, but visual or latent prediction alone does not determine whether an imagined future satisfies task-relevant events.
arXiv:2608.24044v1 Announce Type: new Abstract: Latent world models plan by predicting how candidate actions transform learned representations. In self-predictive models, however, the encoder and pre...
The paper introduces an event‑anchored evaluation protocol for video world models, using 62 free‑fall recordings and 124 clips with detailed release and impact annotations. It finds that while some models (Runway, Veo) can generate release and impact events with high accuracy, they often start them late, and others (Cosmos‑Predict‑2.5, MAGI‑1) rarely produce measurable consequences. A human study shows that people’s predictions align with recorded futures but also reveal ambiguity in plausible continuations, highlighting that physical foresight requires initiating, timing, and realizing motion correctly.
arXiv:2607. 27017v1 Announce Type: new Abstract: A central premise of latent world models is that predicting the future forces a representation to internalize the physics of its environment.
ForeTime‑VLA is a causal vision‑language‑action policy that distills future‑aware representations from a frozen Fast‑WAM teacher, enabling it to anticipate contact events during conveyor‑belt manipulation. The method compresses current and future video latents into a 64‑dimensional target, uses an eight‑frame history encoder to predict this target along with manipulation phase and time‑to‑transition, and conditions a VLM prefix on future tokens and phase. On a deduplicated conveyor‑belt dataset, ForeTime‑VLA reduces test MAE by 2.63% and L2 by 3.02%, while real‑robot experiments show significantly higher grasp success rates compared to the next‑best reference. whyItMatters":"The approach demonstrates that distilling future‑token knowledge from a world‑action model can improve dynamic manipulation performance without the computational cost of running the teacher at inference time."
arXiv:2608. 12939v1 Announce Type: new Abstract: Joint-embedding predictive architectures (JEPAs) learn world models that predict in a compact latent space rather than in pixels, reducing the pressure to model nuisance appearance.
arXiv:2606. 17572v1 Announce Type: new Abstract: Learned dynamics models often answer global physical questions, such as fault severity or impact stiffness, by pooling a per-step feature sequence into one readout vector.
JEPA-x is a cross‑predictive physics grounding method that aligns visual latent dynamics with privileged physical trajectories. By treating visual observations and physical states as two views of the same action‑conditioned trajectory and sharing a predictor, it forces the model to learn a common transition rule for both modalities. The physical branch is only used during training, so deployment incurs no extra cost, and the approach significantly reduces rollout drift and boosts control success across a multi‑task suite.
arXiv:2608. 15483v1 Announce Type: new Abstract: Modern deep networks are trained through long update trajectories, yet their temporal organization remains less systematically characterized than architectures, losses, or optimizers.
arXiv:2608. 09876v1 Announce Type: cross Abstract: Physically consistent motion planning remains a fundamental challenge in embodied AI, as generated trajectories must strictly conform to real-world execution dynamics.