Saved in:
Bibliographic Details
Main Authors: Wang, Peiyao, Wang, Weining, Li, Qi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.03997
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908632808947712
author Wang, Peiyao
Wang, Weining
Li, Qi
author_facet Wang, Peiyao
Wang, Weining
Li, Qi
contents Recent advances in text-to-video generation have achieved impressive perceptual quality, yet generated content often violates fundamental principles of physical plausibility - manifesting as implausible object dynamics, incoherent interactions, and unrealistic motion patterns. Such failures hinder the deployment of video generation models in embodied AI, robotics, and simulation-intensive domains. To bridge this gap, we propose PhysCorr, a unified framework for modeling, evaluating, and optimizing physical consistency in video generation. Specifically, we introduce PhysicsRM, the first dual-dimensional reward model that quantifies both intra-object stability and inter-object interactions. On this foundation, we develop PhyDPO, a novel direct preference optimization pipeline that leverages contrastive feedback and physics-aware reweighting to guide generation toward physically coherent outputs. Our approach is model-agnostic and scalable, enabling seamless integration into a wide range of video diffusion and transformer-based backbones. Extensive experiments across multiple benchmarks demonstrate that PhysCorr achieves significant improvements in physical realism while preserving visual fidelity and semantic alignment. This work takes a critical step toward physically grounded and trustworthy video generation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_03997
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PhysCorr: Dual-Reward DPO for Physics-Constrained Text-to-Video Generation with Automated Preference Selection
Wang, Peiyao
Wang, Weining
Li, Qi
Computer Vision and Pattern Recognition
Recent advances in text-to-video generation have achieved impressive perceptual quality, yet generated content often violates fundamental principles of physical plausibility - manifesting as implausible object dynamics, incoherent interactions, and unrealistic motion patterns. Such failures hinder the deployment of video generation models in embodied AI, robotics, and simulation-intensive domains. To bridge this gap, we propose PhysCorr, a unified framework for modeling, evaluating, and optimizing physical consistency in video generation. Specifically, we introduce PhysicsRM, the first dual-dimensional reward model that quantifies both intra-object stability and inter-object interactions. On this foundation, we develop PhyDPO, a novel direct preference optimization pipeline that leverages contrastive feedback and physics-aware reweighting to guide generation toward physically coherent outputs. Our approach is model-agnostic and scalable, enabling seamless integration into a wide range of video diffusion and transformer-based backbones. Extensive experiments across multiple benchmarks demonstrate that PhysCorr achieves significant improvements in physical realism while preserving visual fidelity and semantic alignment. This work takes a critical step toward physically grounded and trustworthy video generation.
title PhysCorr: Dual-Reward DPO for Physics-Constrained Text-to-Video Generation with Automated Preference Selection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.03997