ACWM-Phys: Investigating Generalized Physical Interaction in Action-Conditioned Video World Models

• URL: http://arxiv.org/abs/2605.08567v2
• Date: Sat, 16 May 2026 00:08:27 GMT
• Title: ACWM-Phys: Investigating Generalized Physical Interaction in Action-Conditioned Video World Models
• Abstract summary: Action-conditioned world models (ACWMs) have shown strong promise for video prediction and decision-making.<n>We introduce ACWM-Phys, a new benchmark for evaluating action-conditioned prediction under diverse physical dynamics.
• Score: 30.527810700174488
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Action-conditioned world models (ACWMs) have shown strong promise for video prediction and decision-making. However, existing benchmarks are largely restricted to egocentric navigation or narrow, task-specific robotics datasets, offering only limited coverage of the rich physical interactions required for generalized world understanding. We introduce ACWM-Phys, a new benchmark for evaluating action-conditioned prediction under diverse physical dynamics in a clean, controllable simulation environment with a carefully designed action space. ACWM-Phys contains training and evaluation data spanning rigid-body dynamics, kinematics, deformable-object interactions, and particle dynamics. To evaluate both interpolation and generalization, we design in-distribution and out-of-distribution protocols with controlled shifts in interaction patterns or scene configurations. By building the benchmark in a fully controllable simulator, ACWM-Phys enables precise data collection, reproducible evaluation, and systematic analysis of model capabilities for physically grounded world modeling. Through systematic experiments on ACWM-DiT, we find that OoD generalization depends not only on the physical regime but also on effective task complexity: models generalize well on visually simple, low-dimensional interactions with clear geometric structure, but suffer larger drops on deformable contacts, high-dimensional control, and complex articulated motion. This suggests that the model still relies heavily on visual appearance patterns instead of fully learning the underlying physics. Ablations show that cross-attention improves high-dimensional action conditioning, causal VAEs outperform frame-wise encoders, and larger action spaces are harder to model but can improve generalization by providing richer control signals. These findings guide the design of physically grounded world models.

Related papers

• GAUGE: A Measurement-Grounded Benchmark for Physical Fidelity in Simulation Engines and Video World Models [66.77745044586173]
  GAUGE is a real-world-grounded diagnostic benchmark for evaluating how numerical simulators and generative video world models reproduce or deviate from real-world physics.<n>We benchmark Isaac Sim, Genesis, and Newton on 14 task families using generalized trajectory errors, and evaluate 6 image-to-video models on 5 rigid-body tasks.<n>Our results reveal no uniformly faithful physics engine, with the largest discrepancies arising in impulsive contact, rapid textile motion, and volumetric deformation.
  arXiv  Detail & Related papers  (2026-08-06T12:19:38Z)
• IOI: Decoupling Kinematics and Physics for Interactive World Models [46.3330122411516]
  We propose IOI, a hybrid interactive world model integrating kinematics with learned physical dynamics.<n> IOI introduces explicit kinematic guidance, computing kinematics sequences for accurate motion trajectories.<n> experiments on the RoboTwin benchmark validate IOI across kinematic fidelity, out-of-distribution, and policy evaluation.
  arXiv  Detail & Related papers  (2026-06-22T13:09:34Z)
• LaWM: Least Action World Models for Long-Horizon Physical Consistency from Visual Observations [9.35652906982887]
  We propose Least Action World Models (LaWM) to operationalize the Principle of Least Action in learned visual latent space.<n>LaWM encodes observations into learned generalized coordinates, learns a latent discrete Lagrangian over consecutive latent states.<n>Because the transition is induced by a discrete variational principle, LaWM provides a structure-preserving bias for long-horizon visual prediction.
  arXiv  Detail & Related papers  (2026-05-08T07:03:13Z)
• PhyCo: Learning Controllable Physical Priors for Generative Motion [55.59209981836171]
  We present PhyCo, a framework that introduces continuous, interpretable, and physically grounded control into video generation.<n>Our approach integrates three key components: (i) a large-scale dataset of over 100K photorealistic simulation videos where friction, restitution, deformation, and force are systematically varied across diverse scenarios; (ii) physics-supervised fine-tuning of a pretrained diffusion model using a ControlNet conditioned on pixel-aligned physical property maps; and (iii) VLM-guided reward optimization, where a fine-tuned vision-language model evaluates generated videos with targeted physics queries and provides differentiable feedback.
  arXiv  Detail & Related papers  (2026-04-30T17:53:03Z)
• From Perception to Action: An Interactive Benchmark for Vision Reasoning [51.11355591375073]
  Causal Hierarchy of Actions and Interactions (CHAIN) benchmark designed to evaluate whether models can understand, plan, and execute structured action sequences grounded in physical constraints.<n> CHAIN shifts evaluation from passive perception to active problem solving, spanning tasks such as interlocking mechanical puzzles and 3D stacking and packing.<n>Our results show that top-performing models still struggle to internalize physical structure and causal constraints, often failing to produce reliable long-horizon plans and cannot robustly translate perceived structure into effective actions.
  arXiv  Detail & Related papers  (2026-02-24T15:33:02Z)
• ContactGaussian-WM: Learning Physics-Grounded World Model from Videos [25.368710400385392]
  We propose ContactGaussian-WM, a differentiable physics-grounded rigid-body world model capable of learning intricate physical laws directly from sparse and contact-rich video sequences.<n>Extensive simulations and real-world evaluations demonstrate that ContactGaussian-WM outperforms state-of-the-art methods in learning complex scenarios.
  arXiv  Detail & Related papers  (2026-02-11T16:48:13Z)
• Walk through Paintings: Egocentric World Models from Internet Priors [65.30611174953958]
  We present the Egocentric World Model (EgoWM), a simple, architecture-agnostic method that transforms any pretrained video diffusion model into an action-conditioned world model.<n>Rather than training from scratch, we repurpose the rich world priors of Internet-scale video models and inject motor commands through lightweight conditioning layers.<n>Our approach scales naturally across embodiments and action spaces, ranging from 3-DoF mobile robots to 25-DoF humanoids.
  arXiv  Detail & Related papers  (2026-01-21T18:59:32Z)
• Do-Undo: Generating and Reversing Physical Actions in Vision-Language Models [57.71440995598757]
  We introduce the Do-Undo task and benchmark to address a critical gap in vision-language models.<n>Do-Undo requires models to simulate the outcome of a physical action and then accurately reverse it, reflecting true cause-and-effect in the visual world.
  arXiv  Detail & Related papers  (2025-12-15T18:03:42Z)
• Prismatic World Model: Learning Compositional Dynamics for Planning in Hybrid Systems [38.4555621948915]
  Prismatic World Model (PRISM-WM) is designed to decompose complex hybrid dynamics into composable primitives.<n>PRISM-WM significantly reduces rollout drift by accurately modeling sharp mode transitions in system dynamics.
  arXiv  Detail & Related papers  (2025-12-09T09:40:34Z)
• Pre-Trained Video Generative Models as World Simulators [59.546627730477454]
  We propose Dynamic World Simulation (DWS) to transform pre-trained video generative models into controllable world simulators.<n>To achieve precise alignment between conditioned actions and generated visual changes, we introduce a lightweight, universal action-conditioned module.<n> Experiments demonstrate that DWS can be versatilely applied to both diffusion and autoregressive transformer models.
  arXiv  Detail & Related papers  (2025-02-10T14:49:09Z)
• End-to-End Learning of Hybrid Inverse Dynamics Models for Precise and Compliant Impedance Control [16.88250694156719]
  We present a novel hybrid model formulation that enables us to identify fully physically consistent inertial parameters of a rigid body dynamics model. We compare our approach against state-of-the-art inverse dynamics models on a 7 degree of freedom manipulator.
  arXiv  Detail & Related papers  (2022-05-27T07:39:28Z)
• ACID: Action-Conditional Implicit Visual Dynamics for Deformable Object Manipulation [135.10594078615952]
  We introduce ACID, an action-conditional visual dynamics model for volumetric deformable objects. A benchmark contains over 17,000 action trajectories with six types of plush toys and 78 variants. Our model achieves the best performance in geometry, correspondence, and dynamics predictions.
  arXiv  Detail & Related papers  (2022-03-14T04:56:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.