跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 14 天前AI 評分44

PhysOPSD:以程式碼作為特權資訊,提升影片生成的物理合理性

PhysOPSD: Code as Privilege for Physically Plausible Video Generation

AI 導讀

PhysOPSD 以可執行程式碼作為特權物理上下文,透過在策略自蒸餾提升影片生成模型的物理合理性。PhysOPSD-Engine 將真實物理情境轉換為可驗證的模擬程式,並結合雙層物理特權與受控變化。實驗顯示,使用 3K 樣本,PhysOPSD 在多種模型骨幹上勝過 SFT 和 RL 基線,同時保留一般影片生成品質。

正文

Abstract

PhysOPSD uses executable code as privileged physical context to improve the physical plausibility of video generative models through on-policy self-distillation with dual-level privilege and controlled variations.

While modern video generative models achieve remarkable visual fidelity, they remain noticeably brittle when generating physically plausible dynamics and complex interactions. Targeted post-training is crucial, yet supervised fine-tuning suffers from offline state mismatch, and reinforcement learning is hindered by sparse, ill-defined physical reward functions. On-Policy (Self-)Distillation (OPSD) reconciles on-policy rollouts with dense corrective feedback, shifting the core challenge to designing effective privileged context. However, conventional modalities fall short: text prompts are inherently declarative, whereas real reference videos heavily entangle physical laws with visual appearance without parametric control. To this end, we propose PhysOPSD, introducing executable code as privileged physical context for video generation. Unlike static visual observations, executable code instantiates an editable physical world with intrinsic abstraction and controllability. Powered by PhysOPSD-Engine, we translate real-world physical scenarios into verifiable simulation programs. PhysOPSD then incorporates (i) dual-level physical privilege to bridge the information gap via an intermediate bottleneck, alongside (ii) controlled physical variations under shared rollout noise to expose causal dynamic shifts. Extensive experiments demonstrate that PhysOPSD substantially improves physical plausibility across diverse backbones, outperforming SFT and RL baselines with only 3K samples, while preserving general video generation quality.

View arXiv page View PDF Project page GitHub 1 Add to collection

Get this paper in your agent:

hf papers read 2611.12411

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2611.12411 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2611.12411 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2611.12411 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co