VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models

TL;DR

VideoREPA introduces a novel method to enhance physics understanding in text-to-video models by distilling knowledge from foundation models through token relation alignment, leading to more physically plausible video generation.

Contribution

It is the first REPA approach specifically designed for finetuning T2V models to inject physical knowledge and improve physics consistency in generated videos.

Findings

Significant improvement in physics commonsense benchmarks.

Enhanced physical plausibility in generated videos.

Effective distillation of physics understanding from foundation models.

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/.