SGF+: Decoupling Gradient Flows for Autoregressive Video Generation
cs.CV
Submitted: 2026-10-07
Updated: 2026-10-08
Comments: Project page: https://zihan-su.github.io/self-gradient-forcing-plus
Project page: https://zihan-su.github.io/self-gradient-forcing-plus
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion
- SkyReels-V2: Infinite-length Film Generative Model
- Zing-0.5: Toward Playable Worlds with Real-Time Joint Action and Text Control
- Data-Forcing Distillation: Restoring Diversity and Fidelity in Few-Step Video Generation
- Self-Forcing++: Towards Minute-Scale High-Quality Video Generation
- One-Forcing: Towards Stable One-Step Autoregressive Video Generation
- The Llama 3 Herd of Models
- End-to-End Training for Autoregressive Video Diffusion via Self-Resampling
- Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion
- Mixtral of Experts
- MemRoPE: Training-Free Infinite Video Generation via Evolving Memory Tokens
- FreqForcing: Autoregressive Long Video Generation via Spectral Self-Anchoring
- Causal World Modeling for Robot Control
- Precise Editing and Flexible Referencing for Interactable Worlds
- Rolling Forcing: Autoregressive Long Video Diffusion in Real Time
- Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation
- FACT: Failure-Aware Causal Training for World-Action Models
- Movie Gen: A Cast of Media Foundation Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models