文件夹: Papers/Multimodal-Generation/Acceleration--and--Distillation

2026年5月

LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation

2026年5月

On-Policy Adversarial Flow Distillation for Autoregressive Video Generation

2026年5月

One-Forcing: Towards Stable One-Step Autoregressive Video Generation

2026年5月

PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion

2026年4月

Representation Fréchet Loss for Visual Generation

2026年3月

HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoising

2026年3月

Streaming Autoregressive Video Generation via Diagonal Distillation

2026年2月

FlowCache: Flow Caching for Autoregressive Video Generation

2026年2月

Generative Modeling via Drifting

2025年12月

TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times

2025年11月

Image-Free Timestep Distillation via Continuous-Time Consistency with Trajectory-Sampled Pairs

2025年11月

MotionStream: Real-Time Video Generation with Interactive Motion Controls

2025年10月

rCM: Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency

2025年9月

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer

2025年7月

SADA: Stability-guided Adaptive Diffusion Acceleration

2025年7月

StreamDiT: Real-Time Streaming Text-to-Video Generation

2025年6月

DCM: Dual-Expert Consistency Model for Efficient and High-Quality Video Generation

2025年6月

Radial Attention: O(n log n) Sparse Attention with Energy Decay for Long Video Generation

2025年1月

Diffusion Adversarial Post-Training for One-Step Video Generation

2024年5月

DMD2: Improved Distribution Matching Distillation for Fast Image Synthesis

2023年11月

Paper Notes

探索

文件夹: Papers/Multimodal-Generation/Acceleration--and--Distillation

LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation

On-Policy Adversarial Flow Distillation for Autoregressive Video Generation

One-Forcing: Towards Stable One-Step Autoregressive Video Generation

PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion

Representation Fréchet Loss for Visual Generation

HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoising

Streaming Autoregressive Video Generation via Diagonal Distillation

FlowCache: Flow Caching for Autoregressive Video Generation

Generative Modeling via Drifting

TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times

Image-Free Timestep Distillation via Continuous-Time Consistency with Trajectory-Sampled Pairs

MotionStream: Real-Time Video Generation with Interactive Motion Controls

rCM: Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer

SADA: Stability-guided Adaptive Diffusion Acceleration

StreamDiT: Real-Time Streaming Text-to-Video Generation

DCM: Dual-Expert Consistency Model for Efficient and High-Quality Video Generation

Radial Attention: O(n log n) Sparse Attention with Energy Decay for Long Video Generation

Diffusion Adversarial Post-Training for One-Step Video Generation

DMD2: Improved Distribution Matching Distillation for Fast Image Synthesis

DMD: One-step Diffusion with Distribution Matching Distillation