REVIEW 28 cited by
VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models
read the original abstract
Recent advancements in text-to-video (T2V) diffusion models have enabled high-fidelity and realistic video synthesis. However, current T2V models often struggle to generate physically plausible content due to their limited inherent ability to accurately understand physics. We found that while the representations within T2V models possess some capacity for physics understanding, they lag significantly behind those from recent video self-supervised learning methods. To this end, we propose a novel framework called VideoREPA, which distills physics understanding capability from video understanding foundation models into T2V models by aligning token-level relations. This closes the physics understanding gap and enable more physics-plausible generation. Specifically, we introduce the Token Relation Distillation (TRD) loss, leveraging spatio-temporal alignment to provide soft guidance suitable for finetuning powerful pre-trained T2V models, a critical departure from prior representation alignment (REPA) methods. To our knowledge, VideoREPA is the first REPA method designed for finetuning T2V models and specifically for injecting physical knowledge. Empirical evaluations show that VideoREPA substantially enhances the physics commonsense of baseline method, CogVideoX, achieving significant improvement on relevant benchmarks and demonstrating a strong capacity for generating videos consistent with intuitive physics. More video results are available at https://videorepa.github.io/.
Forward citations
Cited by 28 Pith papers
-
VideoRAE: Taming Video Foundation Models for Generative Modeling via Representation Autoencoders
Frozen video foundation features can be compressed into reconstruction-capable, generation-friendly latents that improve video generation quality and convergence speed.
-
VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization
VLMs formulate differentiable rewards from task-specific rules to enable test-time online LoRA optimization of VGMs, delivering 16.7-point gains on symbolic and general video reasoning benchmarks over VLM-as-solver an...
-
VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization
VLMs act as teachers by deriving differentiable rewards from task rules to adapt VGMs via test-time LoRA optimization, delivering 16.7-point average gains on symbolic and general video reasoning benchmarks.
-
Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models
M²-REPA decouples modality-specific features from diffusion intermediates and aligns them to complementary expert foundation models via a multi-modal alignment loss and modality-specific decoupling regularization for ...
-
Olaf-World: Orienting Latent Actions for Video World Modeling
Latent actions become transferable across visual contexts when aligned to temporal feature differences from a frozen video encoder (SeqΔ-REPA), improving zero-shot action transfer and data-efficient adaptation of vide...
-
AgentHOI: Multi-Agent Reasoning for Human-Object-Interaction Video Generation via Implicit Representation Alignment
AgentHOI generates human-object interaction videos from text plus one human image and one object image, using multi-agent action planning and implicit text-to-motion feature alignment inside a video diffusion model.
-
Enhancing Video Physical Consistency via Role-aware Joint Training and Modality-decoupled Denoising
Role-aware joint training plus modality-decoupled denoising and loss annealing improves physical consistency of pretrained video diffusion models while preserving visual quality.
-
Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
Aura combines VLM meta-queries, T5-teacher alignment, subject-aware RoPE shifts, memory tokens, and a large AIGC-curated dataset to claim SOTA multi-element subject-to-video generation under OpenS2V-Eval Total score.
-
DiffusionBench: On Holistic Evaluation of Diffusion Transformers
NanoGen unifies DiT training on ImageNet and T2I, reveals negative Pearson correlations (-0.377 to -0.580) in method rankings across metrics from 21 models, and motivates DiffusionBench for holistic evaluation.
-
GEM-4D: Geometry-Enhanced Video World Models for Robot Manipulation
GEM-4D improves video world models for robot manipulation by distilling 4D geometric correspondences into training and adding an inverse dynamics module, achieving SOTA geometric consistency and 81% real-world success.
-
Spatial Gram Alignment for Ultra-High-Resolution Image Synthesis
Spatial Gram Alignment aligns internal self-similarities of LDM features with foundation priors to reconcile global structure and fine details in ultra-high-resolution text-to-image synthesis.
-
GeoFlow: Enforcing Implicit Geometric Consistency in Video Generation
GeoFlow adds a geometry-consistency reward based on rigid camera flow and object appearance preservation, integrated via reinforcement fine-tuning to improve geometric coherence in video generation.
-
Improved Baselines with Representation Autoencoders
RAE v2 reaches gFID 1.06 on ImageNet-256 in 80 epochs by combining multi-layer encoder sums, complementary REPA targets, and free guidance via output reparameterization.
-
SARA: Semantically Adaptive Relational Alignment for Video Diffusion Models
SARA introduces semantic saliency to guide relational alignment in video diffusion models, improving text following and motion quality over prior alignment methods.
-
SARA: Semantically Adaptive Relational Alignment for Video Diffusion Models
SARA improves text alignment and motion quality in video diffusion models by routing token-relation distillation supervision to semantically salient pairs using a Stage-1 aligner trained with SAM masks and InfoNCE.
-
Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models
M²-REPA decouples modality-specific features inside a diffusion model and aligns each to its matching expert foundation model via an alignment loss plus a decoupling regularizer, yielding better visual quality and lon...
-
Human Cognition in Machines: A Unified Perspective of World Models
The paper introduces a unified framework for world models that fully incorporates all cognitive functions from Cognitive Architecture Theory, highlights under-researched areas in motivation and meta-cognition, and pro...
-
Under One Sun: Multi-Object Generative Perception of Materials and Illumination
Factorizing video editing into semantic-token anchoring and motion-restoration pre-training produces strong zero-shot and SOTA open-source instruction-guided video edits without heavy external structural priors.
-
Transition Matching Distillation for Fast Video Generation
Splitting a video diffusion model into a fixed feature extractor and a small recurrent flow head lets TMD generate videos in one to two effective steps with better VBench scores than prior distilled models.
-
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.
-
Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling
Geometry Forcing aligns video diffusion representations with geometric foundation model features via angular cosine and scale regression objectives to improve 3D consistency in generated videos.
-
PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation
PhysisForcing applies trajectory and relational alignment losses to DiT features in video models, improving physical plausibility on R-Bench, PAI-Bench, and EZS-Bench while raising closed-loop robotic success rates fr...
-
PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation
PhysRAG curates 7K videos from WISA-80K, builds a physical video database, and injects knowledge via learnable queries into a diffusion model to reach SOTA visual quality and physical compliance on PhyGenBench and VBench.
-
MaineCoon: Pursuing A Real-Time Audio-Visual Social World Model
MaineCoon is presented as the first 22B-parameter real-time streaming audio-visual autoregressive model optimized for social-interactive applications, using novel training techniques and an agentic inference framework.
-
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 visu...
-
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 VideoP...
-
GEM-4D: Geometry-Enhanced Video World Models for Robot Manipulation
GEM-4D is a video world model that injects 4D correspondence supervision to improve geometric consistency and robot manipulation success from 61% to 81%.
-
Aligning Perception, Reasoning, Modeling and Interaction: A Survey on Physical AI
A survey of physical AI that distinguishes theoretical physics reasoning from applied understanding and synthesizes advances in symbolic reasoning, embodied systems, and generative models to advocate for physics-groun...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.