Image de couverture de livre

Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Xue, Shuchen, Xie, Tianyu, Hu, Tianyang, Feng, Zijin, Sun, Jiacheng, Kawaguchi, Kenji, Li, Zhenguo, Ma, Zhi-Ming
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning 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

Variational Autoencoding Discrete Diffusion with Enhanced Dimensional Correlations Modeling
par: Xie, Tianyu, et autres
Publié: (2025)

The Surprising Effectiveness of Skip-Tuning in Diffusion Sampling
par: Ma, Jiajun, et autres
Publié: (2024)

Accelerating Diffusion Sampling with Optimized Time Steps
par: Xue, Shuchen, et autres
Publié: (2024)

Improved Diffusion-based Generative Model with Better Adversarial Robustness
par: Wang, Zekun, et autres
Publié: (2025)

Adding Additional Control to One-Step Diffusion with Joint Distribution Matching
par: Luo, Yihong, et autres
Publié: (2025)

Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models
par: Luo, Weijian, et autres
Publié: (2023)

Masked Diffusion Models as Energy Minimization
par: Chen, Sitong, et autres
Publié: (2025)

SA-Solver: Stochastic Adams Solver for Fast Sampling of Diffusion Models
par: Xue, Shuchen, et autres
Publié: (2023)

Noise Consistency Training: A Native Approach for One-Step Generator in Learning Additional Controls
par: Luo, Yihong, et autres
Publié: (2025)

Reward-Instruct: A Reward-Centric Approach to Fast Photo-Realistic Image Generation
par: Luo, Yihong, et autres
Publié: (2025)

Learning Few-Step Diffusion Models by Trajectory Distribution Matching
par: Luo, Yihong, et autres
Publié: (2025)

MDReID: Modality-Decoupled Learning for Any-to-Any Multi-Modal Object Re-Identification
par: Feng, Yingying, et autres
Publié: (2025)

MaskMed: Decoupled Mask and Class Prediction for Medical Image Segmentation
par: Xie, Bin, et autres
Publié: (2025)

Getting More Juice Out of Your Data: Hard Pair Refinement Enhances Visual-Language Models Without Extra Data
par: Wang, Haonan, et autres
Publié: (2023)

Seg2Any: Open-set Segmentation-Mask-to-Image Generation with Precise Shape and Semantic Control
par: Li, Danfeng, et autres
Publié: (2025)

Benchmarking Segmentation Models with Mask-Preserved Attribute Editing
par: Yin, Zijin, et autres
Publié: (2024)

OmniInsert: Mask-Free Video Insertion of Any Reference via Diffusion Transformer Models
par: Chen, Jinshu, et autres
Publié: (2025)

AnyDesign: Versatile Area Fashion Editing via Mask-Free Diffusion
par: Niu, Yunfang, et autres
Publié: (2024)

Learning Single Index Models with Diffusion Priors
par: Tang, Anqi, et autres
Publié: (2025)

MaskAnyNet: Rethinking Masked Image Regions as Valuable Information in Supervised Learning
par: Hong, Jingshan, et autres
Publié: (2025)

Any-Order Flexible Length Masked Diffusion
par: Kim, Jaeyeon, et autres
Publié: (2025)

Any-to-3D Generation via Hybrid Diffusion Supervision
par: Fan, Yijun, et autres
Publié: (2024)

DDT: Decoupled Diffusion Transformer
par: Wang, Shuai, et autres
Publié: (2025)

SAMTok: Representing Any Mask with Two Words
par: Zhou, Yikang, et autres
Publié: (2026)

Structurally Prune Anything: Any Architecture, Any Framework, Any Time
par: Wang, Xun, et autres
Publié: (2024)

Enhancing Text-to-Image Editing via Hybrid Mask-Informed Fusion
par: Li, Aoxue, et autres
Publié: (2024)

AnyMo: Scaling Any-Modality Conditional Motion Generation with Masked Modeling
par: Li, Yiheng, et autres
Publié: (2026)

Improving Adversarial Robustness via Decoupled Visual Representation Masking
par: Liu, Decheng, et autres
Publié: (2024)

ViewMask-1-to-3: Multi-View Consistent Image Generation via Multimodal Diffusion Models
par: Zhu, Ruishu, et autres
Publié: (2025)

Decoupled Diffusion Sparks Adaptive Scene Generation
par: Zhou, Yunsong, et autres
Publié: (2025)

DKDM: Data-Free Knowledge Distillation for Diffusion Models with Any Architecture
par: Xiang, Qianlong, et autres
Publié: (2024)

VidTwin: Video VAE with Decoupled Structure and Dynamics
par: Wang, Yuchi, et autres
Publié: (2024)

MAGREF: Masked Guidance for Any-Reference Video Generation with Subject Disentanglement
par: Deng, Yufan, et autres
Publié: (2025)

Qihoo-T2X: An Efficient Proxy-Tokenized Diffusion Transformer for Text-to-Any-Task
par: Wang, Jing, et autres
Publié: (2024)

Reinforcing Diffusion Models by Direct Group Preference Optimization
par: Luo, Yihong, et autres
Publié: (2025)

LocationAgent: A Hierarchical Agent for Image Geolocation via Decoupling Strategy and Evidence from Parametric Knowledge
par: Li, Qiujun, et autres
Publié: (2026)

OmniControl: Control Any Joint at Any Time for Human Motion Generation
par: Xie, Yiming, et autres
Publié: (2023)

AnyRefill: A Unified, Data-Efficient Framework for Left-Prompt-Guided Vision Tasks
par: Xie, Ming, et autres
Publié: (2025)

Unpaired Deblurring via Decoupled Diffusion Model
par: Cheng, Junhao, et autres
Publié: (2025)

PAGen: Phase-guided Amplitude Generation for Domain-adaptive Object Detection
par: Du, Shuchen, et autres
Publié: (2025)