Adversarial Gradient Forcing for Robust Few-Step Autoregressive Video Generation
cs.CV
Submitted: 2026-10-10
Updated: 2026-10-10
Code: https://github.com/wuli-art/Adversarial-Gradient-Forcing
Project page: https://wuli-art.github.io/Adversarial-Gradient-Forcing
Sources
- Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
- Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets
- Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion
- VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
- Context Forcing: Consistent Autoregressive Video Generation with Long Context
- Self-Forcing++: Towards Minute-Scale High-Quality Video Generation
- Ca2-VDM: Efficient Autoregressive Video Diffusion Model with Causal Generation and Cache Sharing
- Salt: Self-Consistent Distribution Matching with Cache-Aware Training for Fast Video Generation
- Explaining and Harnessing Adversarial Examples
- End-to-End Training for Autoregressive Video Diffusion via Self-Resampling
- Video Diffusion Models
- Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion
- VBench: Comprehensive Benchmark Suite for Video Generative Models
- VBench++: Comprehensive and Versatile Benchmark Suite for Video Generative Models
- HunyuanVideo: A Systematic Framework For Large Video Generative Models
- Professor Forcing: A New Algorithm for Training Recurrent Networks
- Stable Video Infinity: Infinite-Length Video Generation with Error Recycling
- Ms. Forcing: Efficient Streaming Video Generation with Multi-Scale Patchification and Attention
- Autoregressive Adversarial Post-Training for Real-Time Interactive Video Generation
- OPSD-V: On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators
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