Imagen de Portada

Attend to Not Attended: Structure-then-Detail Token Merging for Post-training DiT Acceleration

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Fang, Haipeng, Tang, Sheng, Cao, Juan, Zhang, Enshuo, Tang, Fan, Lee, Tong-Yee
Formato:	Preprint
Publicado:	2025
Materias:	Computer Vision and Pattern Recognition
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

HeadRouter: A Training-free Image Editing Framework for MM-DiTs by Adaptively Routing Attention Heads
por: Xu, Yu, et al.
Publicado: (2024)

PTQ4DiT: Post-training Quantization for Diffusion Transformers
por: Wu, Junyi, et al.
Publicado: (2024)

Where MLLMs Attend and What They Rely On: Explaining Autoregressive Token Generation
por: Chen, Ruoyu, et al.
Publicado: (2025)

Q-DiT: Accurate Post-Training Quantization for Diffusion Transformers
por: Chen, Lei, et al.
Publicado: (2024)

Reasoning to Attend: Try to Understand How <SEG> Token Works
por: Qian, Rui, et al.
Publicado: (2024)

Locality-Attending Vision Transformer
por: Hajimiri, Sina, et al.
Publicado: (2026)

$Δ$-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers
por: Chen, Pengtao, et al.
Publicado: (2024)

Grouping First, Attending Smartly: Training-Free Acceleration for Diffusion Transformers
por: Ren, Sucheng, et al.
Publicado: (2025)

OUSAC: Optimized Guidance Scheduling with Adaptive Caching for DiT Acceleration
por: Sun, Ruitong, et al.
Publicado: (2025)

U-DiTs: Downsample Tokens in U-Shaped Diffusion Transformers
por: Tian, Yuchuan, et al.
Publicado: (2024)

DiVE: DiT-based Video Generation with Enhanced Control
por: Jiang, Junpeng, et al.
Publicado: (2024)

YOSE: You Only Select Essential Tokens for Efficient DiT-based Video Object Removal
por: Wu, Chenyang, et al.
Publicado: (2026)

HyperMotionX: The Dataset and Benchmark with DiT-Based Pose-Guided Human Image Animation of Complex Motions
por: Xu, Shuolin, et al.
Publicado: (2025)

Q-DiT4SR: Exploration of Detail-Preserving Diffusion Transformer Quantization for Real-World Image Super-Resolution
por: Zhang, Xun, et al.
Publicado: (2026)

FIS-DiT: Breaking the Few-Step Video Inference Barrier via Training-Free Frame Interleaved Sparsity
por: Tang, Jian, et al.
Publicado: (2026)

UniAnimate-DiT: Human Image Animation with Large-Scale Video Diffusion Transformer
por: Wang, Xiang, et al.
Publicado: (2025)

Inf-DiT: Upsampling Any-Resolution Image with Memory-Efficient Diffusion Transformer
por: Yang, Zhuoyi, et al.
Publicado: (2024)

Attend and Enrich: Enhanced Visual Prompt for Zero-Shot Learning
por: Liu, Man, et al.
Publicado: (2024)

Remix-DiT: Mixing Diffusion Transformers for Multi-Expert Denoising
por: Fang, Gongfan, et al.
Publicado: (2024)

LRQ-DiT: Log-Rotation Post-Training Quantization of Diffusion Transformers for Image and Video Generation
por: Yang, Lianwei, et al.
Publicado: (2025)

SparseDiT: Token Sparsification for Efficient Diffusion Transformer
por: Chang, Shuning, et al.
Publicado: (2024)

GATS: Gather-Attend-Scatter
por: Zolna, Konrad, et al.
Publicado: (2024)

Attend to what I say: Highlighting relevant content on slides
por: M, Megha Mariam K, et al.
Publicado: (2026)

DiTalker: A Unified DiT-based Framework for High-Quality and Speaking Styles Controllable Portrait Animation
por: Feng, He, et al.
Publicado: (2025)

DiT4Edit: Diffusion Transformer for Image Editing
por: Feng, Kunyu, et al.
Publicado: (2024)

LaVin-DiT: Large Vision Diffusion Transformer
por: Wang, Zhaoqing, et al.
Publicado: (2024)

DiT-Air: Revisiting the Efficiency of Diffusion Model Architecture Design in Text to Image Generation
por: Chen, Chen, et al.
Publicado: (2025)

Unveiling Redundancy in Diffusion Transformers (DiTs): A Systematic Study
por: Sun, Xibo, et al.
Publicado: (2024)

GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features
por: Sträter, Luc P. J., et al.
Publicado: (2024)

Perturb, Attend, Detect and Localize (PADL): Robust Proactive Image Defense
por: Bartolucci, Filippo, et al.
Publicado: (2024)

RealisDance-DiT: Simple yet Strong Baseline towards Controllable Character Animation in the Wild
por: Zhou, Jingkai, et al.
Publicado: (2025)

GenMask: Adapting DiT for Segmentation via Direct Mask Generation
por: Yang, Yuhuan, et al.
Publicado: (2026)

MaterialPicker: Multi-Modal DiT-Based Material Generation
por: Ma, Xiaohe, et al.
Publicado: (2024)

Insert Anything: Image Insertion via In-Context Editing in DiT
por: Song, Wensong, et al.
Publicado: (2025)

Mamoda2.5: Enhancing Unified Multimodal Model with DiT-MoE
por: Shi, Yangming, et al.
Publicado: (2026)

ADP-DiT: Text-Guided Diffusion Transformer for Brain Image Generation in Alzheimer's Disease Progression
por: Lee, Juneyong, et al.
Publicado: (2026)

Lumina-Video: Efficient and Flexible Video Generation with Multi-scale Next-DiT
por: Liu, Dongyang, et al.
Publicado: (2025)

DiT360: High-Fidelity Panoramic Image Generation via Hybrid Training
por: Feng, Haoran, et al.
Publicado: (2025)

DSV: Exploiting Dynamic Sparsity to Accelerate Large-Scale Video DiT Training
por: Tan, Xin, et al.
Publicado: (2025)

Make-Your-Anchor: A Diffusion-based 2D Avatar Generation Framework
por: Huang, Ziyao, et al.
Publicado: (2024)