Importance-Aware OBS Pruning for Diffusion Models

arXiv:2607.20048 · cs.CV · Submitted 2026-07-22 · Read on arXiv

cs.CV

Submitted: 2026-07-22

Updated: 2026-09-28

Code: https://github.com/hpcaitech/Open-Sora

Terminology

• Sparsity — 199× in this paper · explained in 4 episodes
• Pruning — 146× in this paper · explained in 4 episodes
• Diffusion Model — 53× in this paper · explained in 8 episodes
• Diffusion Models — 48× in this paper · explained in 6 episodes
• Transformer — 18× in this paper · explained in 2 episodes
• Fidelity — 12× in this paper · explained in 2 episodes
• Alignment — 10× in this paper · explained in 1 episode
• Calibration — 8× in this paper · explained in 7 episodes
• Generative Model — 7× in this paper · explained in 2 episodes
• Quantization — 7× in this paper · explained in 12 episodes
• Distillation — 6× in this paper · explained in 2 episodes
• Reconstruction Error — 6× in this paper · explained in 2 episodes
• Classifier-Free Guidance — 5× in this paper
• Generative Models — 5× in this paper · explained in 2 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Consistency — 3× in this paper · explained in 3 episodes
• Training-Free Framework — 3× in this paper · explained in 2 episodes
• Ablation study — 2× in this paper · explained in 1 episode
• Model-Agnostic — 2× in this paper · explained in 2 episodes
• Optimal Brain Surgeon — 2× in this paper
• Robustness — 2× in this paper · explained in 5 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Clustering — 1× in this paper · explained in 2 episodes
• Diffusion Process — 1× in this paper · explained in 2 episodes
• Flow Matching — 1× in this paper · explained in 8 episodes
• Generative Modeling — 1× in this paper · explained in 3 episodes
• Latent Diffusion Model — 1× in this paper · explained in 2 episodes
• Latent Diffusion Models — 1× in this paper · explained in 2 episodes
• Latent Space — 1× in this paper · explained in 7 episodes
• Personalization — 1× in this paper · explained in 2 episodes
• Point Clouds — 1× in this paper · explained in 2 episodes
• Rectified Flow — 1× in this paper · explained in 3 episodes
• Semantic Alignment — 1× in this paper · explained in 3 episodes

Sources

• Classifier-Free Diffusion Guidance
• Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding
• Efficient Scaling of Diffusion Transformers for Text-to-Image Generation
• SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer
• FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space
• Denoising Diffusion Implicit Models
• Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
• GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
• SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations
• eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers
• Make-A-Video: Text-to-Video Generation without Text-Video Data
• Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets
• AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning
• DreamFusion: Text-to-3D using 2D Diffusion
• Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model
• 3DDesigner: Towards Photorealistic 3D Object Generation and Editing with Text-guided Diffusion Models
• Point-E: A System for Generating 3D Point Clouds from Complex Prompts
• DiffWave: A Versatile Diffusion Model for Audio Synthesis
• Wuerstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models
• MobileDiffusion: Instant Text-to-Image Generation on Mobile Devices

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