9 citations · 14 across the 30 of their papers we have counts for
31 papers
AttnRouter: Per-Category Attention Routing for Training-Free Image Editing on MMDiT
Guandong Li, Mengxia Ye
We study training-free image editing on Qwen-Image-Edit-2511, a 60-block multi-modal diffusion transformer (MMDiT) that concatenates noise and source-image tokens within a single a…
PhysEdit: Physically-Consistent Region-Aware Image Editing via Adaptive Spatio-Temporal Reasoning
Guandong Li, Mengxia Ye
Image editing instructions are heterogeneous: a color swap, an object insertion, and a physical-action edit all demand different spatial coverage and different reasoning depth, yet…
Edit Fidelity Field: Semantics-Aware Region Isolation for Training-Free Scene Text Editing
Guandong Li, Mengxia Ye
Scene text editing (STE) has achieved remarkable progress in accurately rendering target text through diffusion-based methods. However, we identify a critical yet overlooked proble…
LayerCache: Exploiting Layer-wise Velocity Heterogeneity for Efficient Flow Matching Inference
Guandong Li
Flow Matching models achieve state-of-the-art image generation quality but incur substantial inference cost due to iterative denoising through large Transformer networks. We observ…
AdaEdit: Adaptive Temporal and Channel Modulation for Flow-Based Image Editing
Guandong Li, Zhaobin Chu
Inversion-based image editing in flow matching models has emerged as a powerful paradigm for training-free, text-guided image manipulation. A central challenge in this paradigm is…
Edit Spillover as a Probe: Do Image Editing Models Implicitly Understand World Relations?
Guandong Li, Zhaobin Chu
Instruction-following image editing models are expected to modify only the specified region while keeping the rest of the image unchanged. However, in practice, we observe a pervas…
Frequency-Aware Error-Bounded Caching for Accelerating Diffusion Transformers
Guandong Li
Diffusion Transformers (DiTs) have emerged as the dominant architecture for high-quality image and video generation, yet their iterative denoising process incurs substantial comput…
Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers
Guandong Li
Training-free control over editing intensity is a critical requirement for diffusion-based image editing models built on the Diffusion Transformer (DiT) architecture. Existing atte…
Inject Where It Matters: Training-Free Spatially-Adaptive Identity Preservation for Text-to-Image Personalization
Guandong Li, Mengxia Ye
Personalized text-to-image generation aims to integrate specific identities into arbitrary contexts. However, existing tuning-free methods typically employ Spatially Uniform Visual…
FastUSP: A Multi-Level Collaborative Acceleration Framework for Distributed Diffusion Model Inference
Guandong Li
Large-scale diffusion models such as FLUX (12B parameters) and Stable Diffusion 3 (8B parameters) require multi-GPU parallelism for efficient inference. Unified Sequence Parallelis…
FlexID: Training-Free Flexible Identity Injection via Intent-Aware Modulation for Text-to-Image Generation
Guandong Li, Yijun Ding
Personalized text-to-image generation aims to seamlessly integrate specific identities into textual descriptions. However, existing training-free methods often rely on rigid visual…
DVI: Disentangling Semantic and Visual Identity for Training-Free Personalized Generation
Guandong Li, Yijun Ding
Recent tuning-free identity customization methods achieve high facial fidelity but often overlook visual context, such as lighting, skin texture, and environmental tone. This limit…
EditIDv2: Editable ID Customization with Data-Lubricated ID Feature Integration for Text-to-Image Generation
Guandong Li, Zhaobin Chu
We propose EditIDv2, a tuning-free solution specifically designed for high-complexity narrative scenes and long text inputs. Existing character editing methods perform well under s…
MVNet: Hyperspectral Remote Sensing Image Classification Based on Hybrid Mamba-Transformer Vision Backbone Architecture
Guandong Li, Mengxia Ye
Hyperspectral image (HSI) classification faces challenges such as high-dimensional data, limited training samples, and spectral redundancy, which often lead to overfitting and insu…
Hyperspectral Image Classification via Transformer-based Spectral-Spatial Attention Decoupling and Adaptive Gating
Guandong Li, Mengxia Ye
Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy,…
Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification
Guandong Li, Mengxia Ye
Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy,…
Expert Kernel Generation Network Driven by Contextual Mapping for Hyperspectral Image Classification
Guandong Li, Mengxia Ye
Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy,…
3D Wavelet Convolutions with Extended Receptive Fields for Hyperspectral Image Classification
Guandong Li, Mengxia Ye
Deep neural networks face numerous challenges in hyperspectral image classification, including high-dimensional data, sparse ground object distributions, and spectral redundancy, w…
Spatial-Geometry Enhanced 3D Dynamic Snake Convolutional Neural Network for Hyperspectral Image Classification
Guandong Li, Mengxia Ye
Deep neural networks face several challenges in hyperspectral image classification, including complex and sparse ground object distributions, small clustered structures, and elonga…
Efficient Dynamic Attention 3D Convolution for Hyperspectral Image Classification
Guandong Li, Mengxia Ye
Deep neural networks face several challenges in hyperspectral image classification, including insufficient utilization of joint spatial-spectral information, gradient vanishing wit…
EditID: Training-Free Editable ID Customization for Text-to-Image Generation
Guandong Li, Zhaobin Chu
We propose EditID, a training-free approach based on the DiT architecture, which achieves highly editable customized IDs for text to image generation. Existing text-to-image models…
Layout Control and Semantic Guidance with Attention Loss Backward for T2I Diffusion Model
Guandong Li
Controllable image generation has always been one of the core demands in image generation, aiming to create images that are both creative and logical while satisfying additional sp…
E-Commerce Inpainting with Mask Guidance in Controlnet for Reducing Overcompletion
Guandong Li
E-commerce image generation has always been one of the core demands in the e-commerce field. The goal is to restore the missing background that matches the main product given. In t…
Training-Free Style Consistent Image Synthesis with Condition and Mask Guidance in E-Commerce
Guandong Li
Generating style-consistent images is a common task in the e-commerce field, and current methods are largely based on diffusion models, which have achieved excellent results. This…