Physics engines facilitate large-scale training and evaluation for embodied intelligence, while generative video world models are emerging as implicit simulators of future states and interactions. However, existing evaluations of physical fidelity are often conducted in isolation and rely heavily on perceptual similarity or human judgments, providing limited insight into which physical principles or parameters are violated.
arXiv:2608. 05948v1 Announce Type: new Abstract: Physics engines facilitate large-scale training and evaluation for embodied intelligence, while generative video world models are emerging as implicit simulators of future states and interactions.
Recent advances in image-to-video generation have improved visual realism, making physically grounded and controllable dynamics an important step toward future world simulation. Current models often generate plausible motion, but it is not reliably governed by explicit physical causes, and instance-level constraints can leak or become entangled in multi-object interactions.
arXiv:2608. 09298v1 Announce Type: cross Abstract: Action-conditioned world models (ACWMs) promise to provide embodied AI with scalable predictive simulators for planning, policy evaluation, and data generation.
arXiv:2606. 28128v1 Announce Type: cross Abstract: Video generation models have emerged as a promising paradigm for embodied world simulation.
arXiv:2608. 15555v1 Announce Type: cross Abstract: Video models are increasingly used to predict what happens next in a scene, yet the metrics commonly used to compare their outputs say little about whether the predicted objects move correctly.
Action-conditioned world models (ACWMs) promise to provide embodied AI with scalable predictive simulators for planning, policy evaluation, and data generation. Realizing this promise requires precise action-conditioned transitions rather than merely plausible outputs.
arXiv:2609.39135v1 Announce Type: new Abstract: Physical reasoning from video requires inferring latent physical properties and dynamics beyond direct observation. Direct VLM inference remains unreli...
PhysPlan is a training‑free guidance framework that enhances video diffusion models by incorporating physical awareness through agentic physics simulation. It uses a vision‑language model to generate a Chain‑of‑Visual‑Thought representation of kinematic trajectories and 3D depth, which then drives an object‑centric test‑time optimization that isolates kinematic changes and locks the passive environment. The framework also employs Kinetic Intensity Profiling to adapt hyperparameters to varying physical deformations, and demonstrates superior performance on PhyGenBench and Physics‑IQ benchmarks compared to existing VDM baselines.
arXiv:2610.01092v1 Announce Type: new Abstract: Video generation models are increasingly being explored as world simulators for embodied planning and learning. To do so effectively, these models must...
arXiv:2606. 11324v1 Announce Type: cross Abstract: We introduce Embodied-R1.
The paper introduces KnockGS, a framework that calibrates the elasticity and density of 3D Gaussian objects by analyzing their dynamic response to known forces. By extracting temporal response features from observed dynamics, KnockGS estimates material scales, freezes them into the simulator, and demonstrates that these estimates remain accurate for unseen interactions. Evaluation shows that KnockGS outperforms alternative methods in parameter recovery and response fidelity across diverse material targets.
arXiv:2607. 20653v1 Announce Type: cross Abstract: Predicting how deformable objects evolve under robotic manipulation is a longstanding challenge.