# Bootstrapping Physics-Grounded Video Generation through VLM-Guided Iterative Self-Refinement

arXiv: https://arxiv.org/abs/2511.20280 · HTML: https://mlsanity.com/paper/2511.20280/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Self-improving MLLMs and synthetic data, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 20 of its 29 references that are in the map.
- First posted on arXiv 25 Nov 2025.

## Placed in

- Synthetic data and self-improvement for LLMs / [Self-improving MLLMs and synthetic data](https://mlsanity.com/synth/self-improving-mllms-and-synthetic-data/)

## Builds on (most cited first, 20 of 20)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformers](https://mlsanity.com/paper/2205.15868/) (2022)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models](https://mlsanity.com/paper/2402.17177/) (2024)
- [MAGI-1: Autoregressive Video Generation at Scale](https://mlsanity.com/paper/2505.13211/) (2025)
- [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)
- [Do generative video models understand physical principles?](https://mlsanity.com/paper/2501.09038/) (2025)
- [VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generation](https://mlsanity.com/paper/2503.06800/) (2025)
- [VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models](https://mlsanity.com/paper/2505.23656/) (2025)
- [WISA: World Simulator Assistant for Physics-Aware Text-to-Video Generation](https://mlsanity.com/paper/2503.08153/) (2025)
- [Exploring the Evolution of Physics Cognition in Video Generation: A Survey](https://mlsanity.com/paper/2503.21765/) (2025)
- [Think Before You Diffuse: Infusing Physical Rules into Video Diffusion](https://mlsanity.com/paper/2505.21653/) (2025)
- [Cross-Frame Representation Alignment for Fine-Tuning Video Diffusion Models](https://mlsanity.com/paper/2506.09229/) (2025)
- [Exploring Structural Degradation in Dense Representations for Self-supervised Learning](https://mlsanity.com/paper/2510.17299/) (2025)

## Built on by (most cited first, 3 of 3)

- [Physics-IQ Verified](https://mlsanity.com/paper/2606.18943/) (2026)
- [Training-Free Composed Video Retrieval via Visual Representation-Guided Video-LLM Reasoning](https://mlsanity.com/paper/2606.02321/) (2026)
- [Physis-Lang: Self-Evolving Language as a Physical Representation for Video World Model](https://mlsanity.com/paper/2609.40358/) (2026)

## Abstract (arXiv)

Recent progress in video generation has led to impressive visual quality, yet current models still struggle to produce results that align with real-world physical principles. To this end, we propose an iterative self-refinement framework that leverages large language models and vision-language models to provide physics-aware guidance for video generation. Specifically, we introduce a multimodal chain-of-thought (MM-CoT) process that refines prompts based on feedback from physical inconsistencies, progressively enhancing generation quality. This method is training-free and plug-and-play, making it readily applicable to a wide range of video generation models. Experiments on the PhyIQ benchmark show that our method improves the Physics-IQ score from 56.31 to 62.38. We hope this work serves as a preliminary exploration of physics-consistent video generation and may offer insights for future research.