Wave Forcing: Towards Speed-of-Light Streaming Video Generation
cs.CV, cs.DC
Submitted: 2026-10-31
Updated: 2026-10-31
Code: https://github.com/Vchitect/VBench
Project page: https://nvlabs.github.io/Sana/Sol-Engine/StreamRender-H3
Terminology
Sources
- Code World Model: Coding Agent as World Brain
- SolarRenderer: Few-Step Generative Reference Rendering
- Live Avatar: Streaming Real-time Audio-Driven Avatar Generation with Infinite Length
- Multimodal Diffusion Forcing for Forceful Manipulation
- DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
- Hybrid Static/Dynamic Schedules for Tiled Polyhedral Programs
- Multi-Block Diffusion Language Models
- TBD-VLA: Temporal Block Diffusion Vision Language Action Model
- Movie Gen: A Cast of Media Foundation Models
- StreamEval: Evaluating Interactive Streaming Video Generation Models
- Can World and Action Modeling Benefit Each Other under Unified Causal Training?
- SGF+: Decoupling Gradient Flows for Autoregressive Video Generation
- Wan: Open and Advanced Large-Scale Video Generative Models
- MemForest: Efficient Agent Memory Management via EventTree Partitioning and Progressive Merging
- BlockVLA: Accelerating Autoregressive VLA via Block Diffusion Finetuning
- Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing
- Adaptive Resource Management and Quality Control for Streaming Video Generation
- Helios: Real Real-Time Long Video Generation Model
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