# PhyRPR: Training-Free Physics-Constrained Video Generation

arXiv: https://arxiv.org/abs/2601.09255 · HTML: https://mlsanity.com/paper/2601.09255/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 19 of its 20 references that are in the map.
- First posted on arXiv 14 Jan 2026.

## Placed in

- Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/)

## Builds on (most cited first, 19 of 19)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations](https://mlsanity.com/paper/2108.01073/) (2021)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [Show-o: One Single Transformer to Unify Multimodal Understanding and Generation](https://mlsanity.com/paper/2408.12528/) (2024)
- [LTX-Video: Realtime Video Latent Diffusion](https://mlsanity.com/paper/2501.00103/) (2024)
- [BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset](https://mlsanity.com/paper/2505.09568/) (2025)
- [World Simulation with Video Foundation Models for Physical AI](https://mlsanity.com/paper/2511.00062/) (2025)
- [Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation](https://mlsanity.com/paper/2410.05363/) (2024)
- [UniVideo: Unified Understanding, Generation, and Editing for Videos](https://mlsanity.com/paper/2510.08377/) (2025)
- [NewtonGen: Physics-Consistent and Controllable Text-to-Video Generation via Neural Newtonian Dynamics](https://mlsanity.com/paper/2509.21309/) (2025)
- [VChain: Chain-of-Visual-Thought for Reasoning in Video Generation](https://mlsanity.com/paper/2510.05094/) (2025)
- [What about gravity in video generation? Post-Training Newton's Laws with Verifiable Rewards](https://mlsanity.com/paper/2512.00425/) (2025)
- [UniVid: The Open-Source Unified Video Model](https://mlsanity.com/paper/2509.24200/) (2025)
- [ProPhy: Progressive Physical Alignment for Dynamic World Simulation](https://mlsanity.com/paper/2512.05564/) (2025)
- [Benchmarking Scientific Understanding and Reasoning for Video Generation using VideoScience-Bench](https://mlsanity.com/paper/2512.02942/) (2025)

## Built on by (most cited first, 7 of 7)

- [MMPhysVideo: Physically Plausible Video Generation Through Joint RGB-Perception Modeling](https://mlsanity.com/paper/2604.02817/) (2026)
- [Image-to-Video Diffusion: From Foundations to Open Frontiers](https://mlsanity.com/paper/2605.17248/) (2026)
- [Physics-Informed Video Generation via Mixture-of-Experts Latent Alignment](https://mlsanity.com/paper/2606.04737/) (2026)
- [DeforM: Reasoning-Guided Physics-Aware Video Generation via Spatial-Temporal Masking](https://mlsanity.com/paper/2607.18664/) (2026)
- [Physics-Aware Video Generation via Agentic Planning and Graph-Guided Optimization](https://mlsanity.com/paper/2609.13006/) (2026)
- [PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation](https://mlsanity.com/paper/2606.26916/) (2026)
- [World Models' Last Exam in Physics](https://mlsanity.com/paper/2610.08791/) (2026)

## Abstract (arXiv)

Recent diffusion-based video generation models can synthesize visually plausible videos, yet they often struggle to satisfy physical constraints. A key reason is that most existing approaches remain single-stage: they entangle high-level physical understanding with low-level visual synthesis, making it hard to generate content that require explicit physical reasoning. To address this limitation, we propose a training-free three-stage pipeline,textitPhyRPR:textitPhyulineReason--textitPhyulinePlan--textitPhyulineRefine, which decouples physical understanding from visual synthesis. Specifically, textitPhyReason uses a large multimodal model for physical state reasoning and an image generator for keyframe synthesis; textitPhyPlan deterministically synthesizes a controllable coarse motion scaffold; and textitPhyRefine injects this scaffold into diffusion sampling via a latent fusion strategy to refine appearance while preserving the planned dynamics. This staged design enables explicit physical control during generation. Extensive experiments under physics constraints show that our method consistently improves physical plausibility and motion controllability.