HybridStitch: Token-Adaptive Model Stitching for Efficient Diffusion Generation

arXiv:2603.07815 · cs.CV, cs.AI · Submitted 2026-03-08 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-03-08

Updated: 2026-10-14

Code: https://github.com/vipshop/cache-dit

Terminology

• Diffusion Model — 1× in this paper · explained in 9 episodes
• Diffusion Models — 1× in this paper · explained in 6 episodes

Sources

• SRDiffusion: Accelerate Video Diffusion Inference via Sketching-Rendering Cooperation
• Scaling Rectified Flow Transformers for High-Resolution Image Synthesis
• xDiT: an Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism
• Distribution Matching Distillation Meets Reinforcement Learning
• FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space
• Microsoft COCO: Common Objects in Context
• Decoupled DMD: CFG Augmentation as the Spear, Distribution Matching as the Shield
• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis
• FlexCache: Flexible Approximate Cache System for Video Diffusion
• Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer
• Qwen-Image Technical Report
• SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning
• SLA2: Sparse-Linear Attention with Learnable Routing and QAT
• SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-Bit Training

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