PhysOPSD: Code as Privilege for Physically Plausible Video Generation
cs.CV
Submitted: 2026-11-09
Updated: 2026-11-09
Code: https://github.com/aigc3d/PhysOPSD
Project page: https://haroldchen19.github.io/PhysOPSD-site
Terminology
Sources
- Cosmos World Foundation Model Platform for Physical AI
- Code2Video: A Code-centric Paradigm for Educational Video Generation
- Flow-OPD: On-Policy Distillation for Flow Matching Models
- NEWTON: Agentic Planning for Physically Grounded Video Generation
- LTX-Video: Realtime Video Latent Diffusion
- PhysMaster: Mastering Physical Representation for Video Generation via Reinforcement Learning
- D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models
- HiPhy: Hierarchical Alignment for Physically-Plausible Multi-Principle Video Generation
- HunyuanVideo: A Systematic Framework For Large Video Generative Models
- What about gravity in video generation? Post-Training Newton's Laws with Verifiable Rewards
- VideoCoCo: Code-as-CoT for Physically-Consistent Video Generation via an Agentic Dual-Engine System
- DiffusionOPD: A Unified Perspective of On-Policy Distillation in Diffusion Models
- LEGO-Anything: Coding Agents for 3D Scene Reconstruction
- Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
- PhyGround: Benchmarking Physical Reasoning in Generative World Models
- Flow Matching for Generative Modeling
- OPSD-V: On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators
- MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training
- OmniWeaving: Towards Unified Video Generation with Free-form Composition and Reasoning
- Privileged Information Distillation for Language Models
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs