PhysRVG：面向视频生成模型的物理感知统一强化学习框架

摘要

物理定律是真实视觉模拟的根基，却在基于Transformer的视频生成领域遭到显著忽视。这一空白凸显了渲染刚体运动——经典力学核心准则——的关键局限性。虽然计算机图形学与物理模拟器能轻松运用牛顿公式建模此类碰撞，现代预训练-微调范式却在像素级全局去噪过程中摒弃了物体刚性的概念。即便完全正确的数学约束，在训练后阶段的模型优化中也被视作次优解（即条件），从根本上限制了生成视频的物理真实感。基于这些考量，我们首次提出面向视频生成模型的物理感知强化学习范式，该范式直接在髙维空间强制执行物理碰撞规则，确保物理知识被严格应用而非作为条件处理。随后，我们将该范式扩展为统一框架Mimicry-Discovery循环（MDcycle），在充分保留模型利用物理基础反馈能力的同时实现大规模微调。为验证方法有效性，我们构建了新基准测试集PhysRVGBench，并通过大量定性与定量实验全面评估其性能。

English

Physical principles are fundamental to realistic visual simulation, but remain a significant oversight in transformer-based video generation. This gap highlights a critical limitation in rendering rigid body motion, a core tenet of classical mechanics. While computer graphics and physics-based simulators can easily model such collisions using Newton formulas, modern pretrain-finetune paradigms discard the concept of object rigidity during pixel-level global denoising. Even perfectly correct mathematical constraints are treated as suboptimal solutions (i.e., conditions) during model optimization in post-training, fundamentally limiting the physical realism of generated videos. Motivated by these considerations, we introduce, for the first time, a physics-aware reinforcement learning paradigm for video generation models that enforces physical collision rules directly in high-dimensional spaces, ensuring the physics knowledge is strictly applied rather than treated as conditions. Subsequently, we extend this paradigm to a unified framework, termed Mimicry-Discovery Cycle (MDcycle), which allows substantial fine-tuning while fully preserving the model's ability to leverage physics-grounded feedback. To validate our approach, we construct new benchmark PhysRVGBench and perform extensive qualitative and quantitative experiments to thoroughly assess its effectiveness.

PhysRVG：面向视频生成模型的物理感知统一强化学习框架

PhysRVG: Physics-Aware Unified Reinforcement Learning for Video Generative Models

摘要

Support