Image de couverture de livre

AttnRouter: Per-Category Attention Routing for Training-Free Image Editing on MMDiT

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Li, Guandong, Ye, Mengxia
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Edit Fidelity Field: Semantics-Aware Region Isolation for Training-Free Scene Text Editing
par: Li, Guandong, et autres
Publié: (2026)

FreeFlux: Understanding and Exploiting Layer-Specific Roles in RoPE-Based MMDiT for Versatile Image Editing
par: Wei, Tianyi, et autres
Publié: (2025)

Inject Where It Matters: Training-Free Spatially-Adaptive Identity Preservation for Text-to-Image Personalization
par: Li, Guandong, et autres
Publié: (2026)

Unraveling MMDiT Blocks: Training-free Analysis and Enhancement of Text-conditioned Diffusion
par: Li, Binglei, et autres
Publié: (2026)

Enhancing MMDiT-Based Text-to-Image Models for Similar Subject Generation
par: Wei, Tianyi, et autres
Publié: (2024)

PhysEdit: Physically-Consistent Region-Aware Image Editing via Adaptive Spatio-Temporal Reasoning
par: Li, Guandong, et autres
Publié: (2026)

Efficient Dynamic Attention 3D Convolution for Hyperspectral Image Classification
par: Li, Guandong, et autres
Publié: (2025)

Hyperspectral Image Classification via Transformer-based Spectral-Spatial Attention Decoupling and Adaptive Gating
par: Li, Guandong, et autres
Publié: (2025)

E-MMDiT: Revisiting Multimodal Diffusion Transformer Design for Fast Image Synthesis under Limited Resources
par: Shen, Tong, et autres
Publié: (2025)

Expert Kernel Generation Network Driven by Contextual Mapping for Hyperspectral Image Classification
par: Li, Guandong, et autres
Publié: (2025)

3D Wavelet Convolutions with Extended Receptive Fields for Hyperspectral Image Classification
par: Li, Guandong, et autres
Publié: (2025)

Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification
par: Li, Guandong, et autres
Publié: (2025)

MVNet: Hyperspectral Remote Sensing Image Classification Based on Hybrid Mamba-Transformer Vision Backbone Architecture
par: Li, Guandong, et autres
Publié: (2025)

Spatial-Geometry Enhanced 3D Dynamic Snake Convolutional Neural Network for Hyperspectral Image Classification
par: Li, Guandong, et autres
Publié: (2025)

Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers
par: Li, Guandong
Publié: (2026)

Training-Free Style Consistent Image Synthesis with Condition and Mask Guidance in E-Commerce
par: Li, Guandong
Publié: (2024)

HeadRouter: A Training-free Image Editing Framework for MM-DiTs by Adaptively Routing Attention Heads
par: Xu, Yu, et autres
Publié: (2024)

EditID: Training-Free Editable ID Customization for Text-to-Image Generation
par: Li, Guandong, et autres
Publié: (2025)

FlexID: Training-Free Flexible Identity Injection via Intent-Aware Modulation for Text-to-Image Generation
par: Li, Guandong, et autres
Publié: (2026)

AdaEdit: Adaptive Temporal and Channel Modulation for Flow-Based Image Editing
par: Li, Guandong, et autres
Publié: (2026)

Edit Spillover as a Probe: Do Image Editing Models Implicitly Understand World Relations?
par: Li, Guandong, et autres
Publié: (2026)

DVI: Disentangling Semantic and Visual Identity for Training-Free Personalized Generation
par: Li, Guandong, et autres
Publié: (2025)

Layout Control and Semantic Guidance with Attention Loss Backward for T2I Diffusion Model
par: Li, Guandong
Publié: (2024)

GeoRouter: Dynamic Paradigm Routing for Worldwide Image Geolocalization
par: Jia, Pengyue, et autres
Publié: (2026)

FusionEdit: Semantic Fusion and Attention Modulation for Training-Free Image Editing
par: Lai, Yongwen, et autres
Publié: (2026)

Self-Attention Decomposition For Training Free Diffusion Editing
par: Anand, Tharun, et autres
Publié: (2025)

D-Attn: Decomposed Attention for Large Vision-and-Language Models
par: Kuo, Chia-Wen, et autres
Publié: (2025)

DiTFastAttn: Attention Compression for Diffusion Transformer Models
par: Yuan, Zhihang, et autres
Publié: (2024)

PerSense: Training-Free Personalized Instance Segmentation in Dense Images
par: Siddiqui, Muhammad Ibraheem, et autres
Publié: (2024)

Training-Free Multi-Concept Image Editing
par: Foteinopoulou, Niki, et autres
Publié: (2026)

TS-Attn: Temporal-wise Separable Attention for Multi-Event Video Generation
par: Zhang, Hongyu, et autres
Publié: (2026)

Reversible Inversion for Training-Free Exemplar-guided Image Editing
par: Li, Yuke, et autres
Publié: (2025)

AttnMod: Attention-Based New Art Styles
par: Su, Shih-Chieh
Publié: (2024)

Towards PerSense++: Advancing Training-Free Personalized Instance Segmentation in Dense Images
par: Siddiqui, Muhammad Ibraheem, et autres
Publié: (2025)

GradAttn: Replacing Fixed Residual Connections with Task-Modulated Attention Pathways
par: Ghoshal, Soudeep, et autres
Publié: (2026)

AFter: Attention-based Fusion Router for RGBT Tracking
par: Lu, Andong, et autres
Publié: (2024)

EditIDv2: Editable ID Customization with Data-Lubricated ID Feature Integration for Text-to-Image Generation
par: Li, Guandong, et autres
Publié: (2025)

Frequency-Aware Error-Bounded Caching for Accelerating Diffusion Transformers
par: Li, Guandong
Publié: (2026)

FastUSP: A Multi-Level Collaborative Acceleration Framework for Distributed Diffusion Model Inference
par: Li, Guandong
Publié: (2026)

E-Commerce Inpainting with Mask Guidance in Controlnet for Reducing Overcompletion
par: Li, Guandong
Publié: (2024)