# PhyCo: Learning Controllable Physical Priors for Generative Motion

arXiv: https://arxiv.org/abs/2604.28169 · HTML: https://mlsanity.com/paper/2604.28169/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 40 of its 45 references that are in the map.
- First posted on arXiv 30 Apr 2026.

## Placed in

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

## Builds on (most cited first, 30 of 40)

- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformers](https://mlsanity.com/paper/2205.15868/) (2022)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [CLEVRER: CoLlision Events for Video REpresentation and Reasoning](https://mlsanity.com/paper/1910.01442/) (2019)
- [Directly Fine-Tuning Diffusion Models on Differentiable Rewards](https://mlsanity.com/paper/2309.17400/) (2023)
- [Analyzing and Improving the Training Dynamics of Diffusion Models](https://mlsanity.com/paper/2312.02696/) (2023)
- [PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics](https://mlsanity.com/paper/2311.12198/) (2023)
- [Aligning Text-to-Image Diffusion Models with Reward Backpropagation](https://mlsanity.com/paper/2310.03739/) (2023)
- [DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory](https://mlsanity.com/paper/2308.08089/) (2023)
- [PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation](https://mlsanity.com/paper/2404.13026/) (2024)
- [Direct-a-Video: Customized Video Generation with User-Directed Camera Movement and Object Motion](https://mlsanity.com/paper/2402.03162/) (2024)
- [CamCo: Camera-Controllable 3D-Consistent Image-to-Video Generation](https://mlsanity.com/paper/2406.02509/) (2024)
- [Do generative video models understand physical principles?](https://mlsanity.com/paper/2501.09038/) (2025)
- [DragAnything: Motion Control for Anything using Entity Representation](https://mlsanity.com/paper/2403.07420/) (2024)
- [Physion: Evaluating Physical Prediction from Vision in Humans and Machines](https://mlsanity.com/paper/2106.08261/) (2021)
- [CamI2V: Camera-Controlled Image-to-Video Diffusion Model](https://mlsanity.com/paper/2410.15957/) (2024)
- [Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise](https://mlsanity.com/paper/2501.08331/) (2025)
- [Boximator: Generating Rich and Controllable Motions for Video Synthesis](https://mlsanity.com/paper/2402.01566/) (2024)
- [Video Diffusion Alignment via Reward Gradients](https://mlsanity.com/paper/2407.08737/) (2024)
- [Reconstruction and Simulation of Elastic Objects with Spring-Mass 3D Gaussians](https://mlsanity.com/paper/2403.09434/) (2024)
- [CoPhy: Counterfactual Learning of Physical Dynamics](https://mlsanity.com/paper/1909.12000/) (2019)
- [SG-I2V: Self-Guided Trajectory Control in Image-to-Video Generation](https://mlsanity.com/paper/2411.04989/) (2024)
- [ComPhy: Compositional Physical Reasoning of Objects and Events from Videos](https://mlsanity.com/paper/2205.01089/) (2022)
- [Fr'echet Video Motion Distance: A Metric for Evaluating Motion Consistency in Videos](https://mlsanity.com/paper/2407.16124/) (2024)
- [ShapeStacks: Learning Vision-Based Physical Intuition for Generalised Object Stacking](https://mlsanity.com/paper/1804.08018/) (2018)
- [Physion++: Evaluating Physical Scene Understanding that Requires Online Inference of Different Physical Properties](https://mlsanity.com/paper/2306.15668/) (2023)

## Built on by (most cited first, 5 of 5)

- [NEWTON: Agentic Planning for Physically Grounded Video Generation](https://mlsanity.com/paper/2605.18396/) (2026)
- [PhysEditWorld: A Large-Scale Dataset Toward Physics-Editable World Models](https://mlsanity.com/paper/2606.26694/) (2026)
- [PhiZero: A World Model Built Around Physical Language](https://mlsanity.com/paper/2607.28624/) (2026)
- [PhysWeep: Does a Video Generator Realize the Physics You Ask For?](https://mlsanity.com/paper/2609.06207/) (2026)
- [VeriPhy: Agentic Physical Reasoning for World Model Evaluation and Refinement](https://mlsanity.com/paper/2609.03153/) (2026)

## Abstract (arXiv)

Modern video diffusion models excel at appearance synthesis but still struggle with physical consistency: objects drift, collisions lack realistic rebound, and material responses seldom match their underlying properties. We present PhyCo, a framework that introduces continuous, interpretable, and physically grounded control into video generation. 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. This combination enables a generative model to produce physically consistent and controllable outputs through variations in physical attributes-without any simulator or geometry reconstruction at inference. On the Physics-IQ benchmark, PhyCo significantly improves physical realism over strong baselines, and human studies confirm clearer and more faithful control over physical attributes. Our results demonstrate a scalable path toward physically consistent, controllable generative video models that generalize beyond synthetic training environments.