A 5B-parameter latent diffusion model generates real-time four-player Rocket League matches conditioned on all players' actions, staying stable far beyond its training horizon.
hub Canonical reference
VideoPhy: Evaluating Physical Commonsense for Video Generation
Canonical reference. 78% of citing Pith papers cite this work as background.
48
Pith papers citing it
Background
78% of classified citations
abstract
Recent advances in internet-scale video data pretraining have led to the development of text-to-video generative models that can create high-quality videos across a broad range of visual concepts, synthesize realistic motions and render complex objects. Hence, these generative models have the potential to become general-purpose simulators of the physical world. However, it is unclear how far we are from this goal with the existing text-to-video generative models. To this end, we present VideoPhy, a benchmark designed to assess whether the generated videos follow physical commonsense for real-world activities (e.g. marbles will roll down when placed on a slanted surface). Specifically, we curate diverse prompts that involve interactions between various material types in the physical world (e.g., solid-solid, solid-fluid, fluid-fluid). We then generate videos conditioned on these captions from diverse state-of-the-art text-to-video generative models, including open models (e.g., CogVideoX) and closed models (e.g., Lumiere, Dream Machine). Our human evaluation reveals that the existing models severely lack the ability to generate videos adhering to the given text prompts, while also lack physical commonsense. Specifically, the best performing model, CogVideoX-5B, generates videos that adhere to the caption and physical laws for 39.6% of the instances. VideoPhy thus highlights that the video generative models are far from accurately simulating the physical world. Finally, we propose an auto-evaluator, VideoCon-Physics, to assess the performance reliably for the newly released models.
hub tools
citation-role summary
background 6
dataset 2
baseline 1
citation-polarity summary
representative citing papers
ScratchWorld benchmark finds that language models achieve at most 13.8% value-aware changed-field F1 on replay-verified Scratch state transitions and frequently ignore executable rules.
None of ten tested video-generation models reliably remembers objects after occlusion in dynamic scenes; static-camera videos inflate consistency scores.
PhaseLock extracts motion priors from 2-step inference and enforces them via Latent Delta Guidance to raise physical consistency scores by 6.2 points on average in image-to-video diffusion models.
Dream.exe evaluates 8 video generation models on 101 manipulation tasks by converting generated videos into executable robot trajectories in a simulator, finding measurable success rates that visual metrics do not predict.
MBench is a new benchmark that quantifies long-term memory in video world models via three hierarchical consistency dimensions evaluated on curated real videos.
YoCausal benchmark shows video diffusion models detect the arrow of time but lack genuine causal understanding relative to humans.
What-If World is a new paired-prompt benchmark showing that nine state-of-the-art video generation models achieve at most 52% on causal intervention tests and cluster near 28% for open-source systems.
WBench is a benchmark with 289 test cases and 1,058 turns for evaluating interactive world models using 22 automated metrics validated against human judgments.
CRONOS benchmark shows recent open-source video generators fail to preserve physical consistency under controlled changes to viewpoint, scene, object category, and appearance.
MechVerse benchmark shows current video generation models preserve appearance but fail at mechanically admissible motion, with errors rising as coupling complexity increases.
PhyGround is a new benchmark with curated prompts, a 13-law taxonomy, large-scale human annotations, and an open physics-specialized VLM judge for evaluating physical reasoning in generative video models.
AV-Phys Bench shows that current joint audio-video models lack robust physical commonsense, with major drops on transitions and deliberate anti-physics prompts.
CMTA detects AI-generated videos by capturing unnatural temporal stability in visual-textual semantic alignment via joint embeddings and multi-grained temporal modeling, outperforming prior methods in cross-generator tests.
MoRight disentangles object and camera motion via canonical-view specification and temporal cross-view attention, while decomposing motion into active user-driven and passive consequence components to learn and apply causality in video generation.
PlayWorld learns high-fidelity robot world models from unsupervised self-play, producing physically consistent video predictions that outperform models trained on human data and enabling 65% better real-world policy performance via model-based RL.
VideoASMR-Bench shows state-of-the-art VLMs fail to reliably detect AI-generated ASMR videos from real ones, though humans can still identify the fakes relatively easily.
DreamGen trains robot policies on synthetic trajectories from adapted video world models, enabling a humanoid robot to perform 22 new behaviors in seen and unseen environments from a single pick-and-place teleoperation dataset.
CrashTwin recovers metric-scale crash dynamics from monocular rollouts and shows that strong visual scores routinely mask large momentum, energy, and identity violations in world models.
SimWorlds presents a multi-agent system with planner-coder-reviewer workflow, layered scene protocol, and runtime inspection tools to create dynamic 4D scenes from text, plus the 4DBuildBench benchmark showing outperformance over baselines.
EffectivePresentationScorer evaluates paper-to-video talks for instructional quality by checking clear explanation of ideas, prerequisite concepts, and links to contributions, finding that current systems cover topics but fail to teach.
GEOPHYS defines five geometric properties of per-frame embeddings from image encoders that detect physical implausibility in videos with SOTA accuracy and serve as an efficient verifier.
Introduces VideoWeaver benchmark (16 categories, 285 cases) plus agent-as-judge and skill-evolution algorithm to assess and improve agentic long video generation across frameworks.
LaMo adds self-supervised latent motion priors via a motion drift loss during training and motion prior guidance during sampling to boost physical fidelity in video diffusion models like CogVideoX.
citing papers explorer
-
Multiplayer Interactive World Models with Representation Autoencoders
A 5B-parameter latent diffusion model generates real-time four-player Rocket League matches conditioned on all players' actions, staying stable far beyond its training horizon.
-
ScratchWorld: Evaluating If World Models Compute Executable Consequences
ScratchWorld benchmark finds that language models achieve at most 13.8% value-aware changed-field F1 on replay-verified Scratch state transitions and frequently ignore executable rules.
-
MemoBench: Benchmarking World Modeling in Dynamically Changing Environments
None of ten tested video-generation models reliably remembers objects after occlusion in dynamic scenes; static-camera videos inflate consistency scores.
-
Physics in 2-Steps: Locking Motion Priors Before Visual Refinement Erases Them
PhaseLock extracts motion priors from 2-step inference and enforces them via Latent Delta Guidance to raise physical consistency scores by 6.2 points on average in image-to-video diffusion models.
-
Dream.exe: Can Video Generation Models Dream Executable Robot Manipulation?
Dream.exe evaluates 8 video generation models on 101 manipulation tasks by converting generated videos into executable robot trajectories in a simulator, finding measurable success rates that visual metrics do not predict.
-
MBench: A Comprehensive Benchmark on Memory Capability for Video World Models
MBench is a new benchmark that quantifies long-term memory in video world models via three hierarchical consistency dimensions evaluated on curated real videos.
-
YoCausal: How Far is Video Generation from World Model? A Causality Perspective
YoCausal benchmark shows video diffusion models detect the arrow of time but lack genuine causal understanding relative to humans.
-
What-If World: A Causal Benchmark for General World Models in Embodied Scenarios
What-If World is a new paired-prompt benchmark showing that nine state-of-the-art video generation models achieve at most 52% on causal intervention tests and cluster near 28% for open-source systems.
-
WBench: A Comprehensive Multi-turn Benchmark for Interactive Video World Model Evaluation
WBench is a benchmark with 289 test cases and 1,058 turns for evaluating interactive world models using 22 automated metrics validated against human judgments.
-
CRONOS: Benchmarking Counterfactual Physical Consistency in Video Models
CRONOS benchmark shows recent open-source video generators fail to preserve physical consistency under controlled changes to viewpoint, scene, object category, and appearance.
-
MechVerse: Evaluating Physical Motion Consistency in Video Generation Models
MechVerse benchmark shows current video generation models preserve appearance but fail at mechanically admissible motion, with errors rising as coupling complexity increases.
-
PhyGround: Benchmarking Physical Reasoning in Generative World Models
PhyGround is a new benchmark with curated prompts, a 13-law taxonomy, large-scale human annotations, and an open physics-specialized VLM judge for evaluating physical reasoning in generative video models.
-
Do Joint Audio-Video Generation Models Understand Physics?
AV-Phys Bench shows that current joint audio-video models lack robust physical commonsense, with major drops on transitions and deliberate anti-physics prompts.
-
CMTA: Leveraging Cross-Modal Temporal Artifacts for Generalizable AI-Generated Video Detection
CMTA detects AI-generated videos by capturing unnatural temporal stability in visual-textual semantic alignment via joint embeddings and multi-grained temporal modeling, outperforming prior methods in cross-generator tests.
-
MoRight: Motion Control Done Right
MoRight disentangles object and camera motion via canonical-view specification and temporal cross-view attention, while decomposing motion into active user-driven and passive consequence components to learn and apply causality in video generation.
-
PlayWorld: Learning Robot World Models from Autonomous Play
PlayWorld learns high-fidelity robot world models from unsupervised self-play, producing physically consistent video predictions that outperform models trained on human data and enabling 65% better real-world policy performance via model-based RL.
-
VideoASMR-Bench: Can AI-Generated ASMR Videos Fool VLMs and Humans?
VideoASMR-Bench shows state-of-the-art VLMs fail to reliably detect AI-generated ASMR videos from real ones, though humans can still identify the fakes relatively easily.
-
DreamGen: Unlocking Generalization in Robot Learning through Video World Models
DreamGen trains robot policies on synthetic trajectories from adapted video world models, enabling a humanoid robot to perform 22 new behaviors in seen and unseen environments from a single pick-and-place teleoperation dataset.
-
A Physics-Grounded Benchmark for Multi-Agent Dynamics in World Models
CrashTwin recovers metric-scale crash dynamics from monocular rollouts and shows that strong visual scores routinely mask large momentum, energy, and identity violations in world models.
-
SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation
SimWorlds presents a multi-agent system with planner-coder-reviewer workflow, layered scene protocol, and runtime inspection tools to create dynamic 4D scenes from text, plus the 4DBuildBench benchmark showing outperformance over baselines.
-
A Good Talk Does not Look Like a Summary, It Teaches You! Measuring Takeaways from Paper-to-Video Talks
EffectivePresentationScorer evaluates paper-to-video talks for instructional quality by checking clear explanation of ideas, prerequisite concepts, and links to contributions, finding that current systems cover topics but fail to teach.
-
GEOPHYS: The Geometry of Physical Plausibility
GEOPHYS defines five geometric properties of per-frame embeddings from image encoders that detect physical implausibility in videos with SOTA accuracy and serve as an efficient verifier.
-
VideoWeaver: Evaluating and Evolving Skills for Agentic Long Video Generation
Introduces VideoWeaver benchmark (16 categories, 285 cases) plus agent-as-judge and skill-evolution algorithm to assess and improve agentic long video generation across frameworks.
-
LaMo: Self-Supervised Latent Motion Priors for Physical Realism in Video Generation
LaMo adds self-supervised latent motion priors via a motion drift loss during training and motion prior guidance during sampling to boost physical fidelity in video diffusion models like CogVideoX.
-
NEWTON: Agentic Planning for Physically Grounded Video Generation
NEWTON improves physical accuracy in video generation by deploying a trainable planner that coordinates physics-aware tools and a verifier, raising joint accuracy on VideoPhy-2 without altering the base generators.
-
Enhancing Train-Free Infinite-Frame Generation for Consistent Long Videos
MIGA introduces two-stage alignment to close train-inference gaps and dual consistency enhancement via self-reflection and long-range guidance to achieve SOTA temporal consistency in infinite-frame video generation on VBench and NarrLV.
-
PanoWorld: Geometry-Consistent Panoramic Video World Modeling
PanoWorld adds depth consistency and trajectory consistency losses plus spherical adaptations to a pre-trained video model, plus a new PanoGeo dataset, to produce geometry-consistent 360 video.
-
Quantitative Video World Model Evaluation for Geometric-Consistency
PDI-Bench computes 3D projective residuals from segmented and tracked points to quantify geometric inconsistency in AI-generated videos.
-
How Far Are Video Models from True Multimodal Reasoning?
Current video models succeed on basic understanding but achieve under 25% success on logically grounded generation and near 0% on interactive generation, exposing gaps in multimodal reasoning.
-
ATSS: Detecting AI-Generated Videos via Anomalous Temporal Self-Similarity
ATSS detects AI-generated videos by measuring unnatural repetitive temporal correlations in triple similarity matrices derived from frame visuals and semantic descriptions.
-
VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction
VisPhyWorld evaluates MLLMs' physical reasoning via executable code generation for video reconstruction, with VisPhyBench showing strong semantics but weak parameter inference and dynamics simulation.
-
PhyDetEx: Detecting and Explaining the Physical Plausibility of T2V Models
A new dataset and fine-tuned VLM detector/explainer called PhyDetEx shows that current T2V models still struggle to generate videos that obey physical laws, with open-source models performing worse.
-
SURF: Signature-Retained Fast Video Generation
SURF accelerates high-resolution video generation up to 12.5x by using noise reshifting for low-res previews from pretrained models and a shifting-window Refiner for efficient upscaling that retains original signatures.
-
RAPO++: Cross-Stage Prompt Optimization for Text-to-Video Generation via Data Alignment and Test-Time Scaling
RAPO++ is a three-stage prompt optimization framework combining retrieval-augmented refinement, closed-loop test-time scaling, and LLM fine-tuning to enhance text-to-video generation quality.
-
Robotic Manipulation by Imitating Generated Videos Without Physical Demonstrations
RIGVid shows that filtered AI-generated videos can serve as effective supervision for complex robotic manipulation tasks without any real demonstrations.
-
MAGI-1: Autoregressive Video Generation at Scale
MAGI-1 is a 24B-parameter autoregressive video world model that predicts denoised frame chunks sequentially with increasing noise to enable causal, scalable, streaming generation up to 4M token contexts.
-
Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation
PhyGenBench supplies 160 prompts across 27 physical laws and an automated LLM/VLM evaluation pipeline to measure physical commonsense compliance in current text-to-video models.
-
Your Data Manifold is Secretly a Reward Model: Shell-LCC for Text-to-Video Generation
Generated video latents are pushed toward a shell-shaped manifold fitted to high-quality SFT video patches, producing a dense reward that reduces blur, over-smoothing, and motion artifacts in text-to-video models.
-
Physics-IQ Verified
Physics-IQ Verified refines 57.6% of samples and 34.8% of prompts from the original benchmark and produces moderate ranking shifts (Kendall's τ = 0.46) across six image-to-video models.
-
Kairos: A Regret-Aware Native World-Action Model Stack for Physical AI
Kairos learns and maintains control-sufficient world states via a cross-embodiment curriculum, hybrid linear temporal attention, and deployment-aware co-design for Physical AI.
-
CP4D: Compositional Physics-aware 4D Scene Generation
CP4D generates physically consistent 4D scenes via compositional integration of pre-trained 3D models, hybrid simulator-diffusion motion synthesis, and automated scene composition.
-
Physics-Informed Video Generation via Mixture-of-Experts Latent Alignment
PILA aligns frozen flow-matching video models to a physics attribute bank via MoE experts and operational residuals, reporting SOTA physical plausibility on VBench-2.0, VideoPhy-2 and PhyGenBench while preserving visual quality.
-
MPMWorlds: Material-Point-Method Simulations for Inferring and Extrapolating Physical Dynamics
Assembles MPM simulation dataset and compares code generation versus video diffusion for inferring physical parameters and extrapolating dynamics from videos.
-
Tempered Self-Similarity Alignment for Physically Plausible Video Generation
Tempered Self-similarity Alignment transfers relational structure from foundation-model STSS into video generators via probabilistic correspondence alignment, yielding reported gains in physical plausibility on VideoPhy benchmarks.
-
DriveCtrl: Conditioned Sim-to-Real Driving Video Generation
DriveCtrl is a depth-conditioned controllable framework that generates realistic driving videos from simulation while preserving annotations and scene dynamics.
-
WorldString: Actionable World Representation
Proposes WorldString, a differentiable neural model for the state manifold of actionable physical objects learned directly from 3D or video data as a building block for world models.
-
World Action Models: The Next Frontier in Embodied AI
The paper introduces World Action Models as a new paradigm unifying predictive world modeling with action generation in embodied foundation models and provides a taxonomy of existing approaches.
-
World Simulation with Video Foundation Models for Physical AI
Cosmos-Predict2.5 unifies text-to-world, image-to-world, and video-to-world generation in one model trained on 200M clips with RL post-training, delivering improved quality and control for physical AI.