# PISA Experiments: Exploring Physics Post-Training for Video Diffusion Models by Watching Stuff Drop

arXiv: https://arxiv.org/abs/2503.09595 · HTML: https://mlsanity.com/paper/2503.09595/

## Key facts

- Cited by 50 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 26 of its 44 references that are in the map.
- First posted on arXiv 12 Mar 2025.

## Placed in

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

## Builds on (most cited first, 26 of 26)

- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [Dream to Control: Learning Behaviors by Latent Imagination](https://mlsanity.com/paper/1912.01603/) (2019)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [Mastering Diverse Domains through World Models](https://mlsanity.com/paper/2301.04104/) (2023)
- [Towards Accurate Generative Models of Video: A New Metric & Challenges](https://mlsanity.com/paper/1812.01717/) (2018)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [Recurrent World Models Facilitate Policy Evolution](https://mlsanity.com/paper/1809.01999/) (2018)
- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [Google Scanned Objects: A High-Quality Dataset of 3D Scanned Household Items](https://mlsanity.com/paper/2204.11918/) (2022)
- [DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors](https://mlsanity.com/paper/2310.12190/) (2023)
- [Learning Interactive Real-World Simulators](https://mlsanity.com/paper/2310.06114/) (2023)
- [Unsupervised Learning of Video Representations using LSTMs](https://mlsanity.com/paper/1502.04681/) (2015)
- [Kubric: A scalable dataset generator](https://mlsanity.com/paper/2203.03570/) (2022)
- [Pyramidal Flow Matching for Efficient Video Generative Modeling](https://mlsanity.com/paper/2410.05954/) (2024)
- [VideoPhy: Evaluating Physical Commonsense for Video Generation](https://mlsanity.com/paper/2406.03520/) (2024)
- [How Far is Video Generation from World Model: A Physical Law Perspective](https://mlsanity.com/paper/2411.02385/) (2024)
- [Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation](https://mlsanity.com/paper/2410.05363/) (2024)
- [Physion: Evaluating Physical Prediction from Vision in Humans and Machines](https://mlsanity.com/paper/2106.08261/) (2021)
- [Video Diffusion Alignment via Reward Gradients](https://mlsanity.com/paper/2407.08737/) (2024)
- [Video as the New Language for Real-World Decision Making](https://mlsanity.com/paper/2402.17139/) (2024)
- [Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks](https://mlsanity.com/paper/1607.02586/) (2016)

## Built on by (most cited first, 30 of 50)

- [Force Prompting: Video Generation Models Can Learn and Generalize Physics-based Control Signals](https://mlsanity.com/paper/2505.19386/) (2025)
- [Solaris: Building a Multiplayer Video World Model in Minecraft](https://mlsanity.com/paper/2602.22208/) (2026)
- [NewtonGen: Physics-Consistent and Controllable Text-to-Video Generation via Neural Newtonian Dynamics](https://mlsanity.com/paper/2509.21309/) (2025)
- [Exploring the Evolution of Physics Cognition in Video Generation: A Survey](https://mlsanity.com/paper/2503.21765/) (2025)
- [Epipolar Geometry Improves Video Generation Models](https://mlsanity.com/paper/2510.21615/) (2025)
- [What about gravity in video generation? Post-Training Newton's Laws with Verifiable Rewards](https://mlsanity.com/paper/2512.00425/) (2025)
- [VGGRPO: Towards World-Consistent Video Generation with 4D Latent Reward](https://mlsanity.com/paper/2603.26599/) (2026)
- [Hierarchical Fine-grained Preference Optimization for Physically Plausible Video Generation](https://mlsanity.com/paper/2508.10858/) (2025)
- [TiViBench: Benchmarking Think-in-Video Reasoning for Video Generative Models](https://mlsanity.com/paper/2511.13704/) (2025)
- [PhysRVG: Physics-Aware Unified Reinforcement Learning for Video Generative Models](https://mlsanity.com/paper/2601.11087/) (2026)
- [Simulating the Visual World with Artificial Intelligence: A Roadmap](https://mlsanity.com/paper/2511.08585/) (2025)
- [Self-Refining Video Sampling](https://mlsanity.com/paper/2601.18577/) (2026)
- [Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players](https://mlsanity.com/paper/2605.28816/) (2026)
- [Physical Simulator In-the-Loop Video Generation](https://mlsanity.com/paper/2603.06408/) (2026)
- [Enhancing Physical Plausibility in Video Generation by Reasoning the Implausibility](https://mlsanity.com/paper/2509.24702/) (2025)
- [PhysMaster: Mastering Physical Representation for Video Generation via Reinforcement Learning](https://mlsanity.com/paper/2510.13809/) (2025)
- [WorldBench: Benchmarking Physical Understanding of World Models by Isolating Physics Concepts](https://mlsanity.com/paper/2601.21282/) (2026)
- [ProPhy: Progressive Physical Alignment for Dynamic World Simulation](https://mlsanity.com/paper/2512.05564/) (2025)
- [4DWorldBench: A Comprehensive Evaluation Framework for 3D/4D World Generation Models](https://mlsanity.com/paper/2511.19836/) (2025)
- [ACWM-Phys: Investigating Generalized Physical Interaction in Action-Conditioned Video World Models](https://mlsanity.com/paper/2605.08567/) (2026)
- [Goal Force: Teaching Video Models To Accomplish Physics-Conditioned Goals](https://mlsanity.com/paper/2601.05848/) (2026)
- [PhysInOne: Visual Physics Learning and Reasoning in One Suite](https://mlsanity.com/paper/2604.09415/) (2026)
- [VideoCoCo: Code-as-CoT for Physically-Consistent Video Generation via an Agentic Dual-Engine System](https://mlsanity.com/paper/2607.27380/) (2026)
- [Planning with Sketch-Guided Verification for Physics-Aware Video Generation](https://mlsanity.com/paper/2511.17450/) (2025)
- [SeeU: Seeing the Unseen World via 4D Dynamics-aware Generation](https://mlsanity.com/paper/2512.03350/) (2025)
- [MoReGen: Multi-Agent Motion-Reasoning Engine for Code-based Text-to-Video Synthesis](https://mlsanity.com/paper/2512.04221/) (2025)
- [PhyCo: Learning Controllable Physical Priors for Generative Motion](https://mlsanity.com/paper/2604.28169/) (2026)
- [GAUGE: A Measurement-Grounded Benchmark for Physical Fidelity in Simulation Engines and Video World Models](https://mlsanity.com/paper/2608.05948/) (2026)
- [RigidBench: Evaluating Rigid-Body Physics in Video Generation Models](https://mlsanity.com/paper/2608.15555/) (2026)
- [Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models](https://mlsanity.com/paper/2510.26782/) (2025)

## Abstract (arXiv)

Large-scale pre-trained video generation models excel in content creation but are not reliable as physically accurate world simulators out of the box. This work studies the process of post-training these models for accurate world modeling through the lens of the simple, yet fundamental, physics task of modeling object freefall. We show state-of-the-art video generation models struggle with this basic task, despite their visually impressive outputs. To remedy this problem, we find that fine-tuning on a relatively small amount of simulated videos is effective in inducing the dropping behavior in the model, and we can further improve results through a novel reward modeling procedure we introduce. Our study also reveals key limitations of post-training in generalization and distribution modeling. Additionally, we release a benchmark for this task that may serve as a useful diagnostic tool for tracking physical accuracy in large-scale video generative model development.