Image de couverture de livre

MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhang, Kewei, Huang, Ye, Deng, Yufan, Yu, Jincheng, Chen, Junsong, Ling, Huan, Xie, Enze, Zhou, Daquan
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

HumanNet: Scaling Human-centric Video Learning to One Million Hours
par: Deng, Yufan, et autres
Publié: (2026)

SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer
par: Zhu, Haoyi, et autres
Publié: (2026)

SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer
par: Xie, Enze, et autres
Publié: (2025)

DC-AR: Efficient Masked Autoregressive Image Generation with Deep Compression Hybrid Tokenizer
par: Wu, Yecheng, et autres
Publié: (2025)

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer
par: Chen, Junsong, et autres
Publié: (2025)

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

SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion Transformer
par: Zhao, Yuyang, et autres
Publié: (2026)

Efficient One-Step Diffusion Restoration Model with Compact Token Compression and Linear Attention
par: Qiao, Bingtian, et autres
Publié: (2026)

Real-time One-Step Diffusion-based Expressive Portrait Videos Generation
par: Guo, Hanzhong, et autres
Publié: (2024)

Interactive Multi-Head Self-Attention with Linear Complexity
par: Kang, Hankyul, et autres
Publié: (2024)

Enhancing Spatial Understanding in Image Generation via Reward Modeling
par: Tang, Zhenyu, et autres
Publié: (2026)

Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation
par: Tang, Lexiang, et autres
Publié: (2025)

SAGA: Selective Adaptive Gating for Efficient and Expressive Linear Attention
par: Cao, Yuan, et autres
Publié: (2025)

PixArt-$α$: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis
par: Chen, Junsong, et autres
Publié: (2023)

TalkCLIP: Talking Head Generation with Text-Guided Expressive Speaking Styles
par: Ma, Yifeng, et autres
Publié: (2023)

Magic 1-For-1: Generating One Minute Video Clips within One Minute
par: Yi, Hongwei, et autres
Publié: (2025)

Selective LoRA for Visual Tokens and Attention Heads
par: Luo, Tiange, et autres
Publié: (2025)

Low-Resolution Self-Attention for Semantic Segmentation
par: Wu, Yu-Huan, et autres
Publié: (2023)

ID-to-3D: Expressive ID-guided 3D Heads via Score Distillation Sampling
par: Babiloni, Francesca, et autres
Publié: (2024)

SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers
par: Xie, Enze, et autres
Publié: (2024)

StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation
par: Zhou, Yupeng, et autres
Publié: (2024)

Restoring Initial Noise Sensitivity in Text-to-Image Distillation via Geometric Alignment
par: Huang, Huayang, et autres
Publié: (2026)

Token-Level Inference-Time Alignment for Vision-Language Models
par: Chen, Kejia, et autres
Publié: (2025)

GLMHA A Guided Low-rank Multi-Head Self-Attention for Efficient Image Restoration and Spectral Reconstruction
par: Ilyas, Zaid, et autres
Publié: (2024)

MoH: Multi-Head Attention as Mixture-of-Head Attention
par: Jin, Peng, et autres
Publié: (2024)

Breaking Complexity Barriers: High-Resolution Image Restoration with Rank Enhanced Linear Attention
par: Ai, Yuang, et autres
Publié: (2025)

FreeBlend: Advancing Concept Blending with Staged Feedback-Driven Interpolation Diffusion
par: Zhou, Yufan, et autres
Publié: (2025)

Exploring Real&Synthetic Dataset and Linear Attention in Image Restoration
par: Du, Yuzhen, et autres
Publié: (2024)

SafePTR: Token-Level Jailbreak Defense in Multimodal LLMs via Prune-then-Restore Mechanism
par: Chen, Beitao, et autres
Publié: (2025)

DC-AE 1.5: Accelerating Diffusion Model Convergence with Structured Latent Space
par: Chen, Junyu, et autres
Publié: (2025)

X-NeMo: Expressive Neural Motion Reenactment via Disentangled Latent Attention
par: Zhao, Xiaochen, et autres
Publié: (2025)

High-Fidelity Mural Restoration via a Unified Hybrid Mask-Aware Transformer
par: Jiang, Jincheng, et autres
Publié: (2026)

Controllable and Expressive One-Shot Video Head Swapping
par: Ji, Chaonan, et autres
Publié: (2025)

Restorer: Removing Multi-Degradation with All-Axis Attention and Prompt Guidance
par: Mao, Jiawei, et autres
Publié: (2024)

Improving Vision Transformers by Overlapping Heads in Multi-Head Self-Attention
par: Zhang, Tianxiao, et autres
Publié: (2024)

Agent Attention: On the Integration of Softmax and Linear Attention
par: Han, Dongchen, et autres
Publié: (2023)

Instant Expressive Gaussian Head Avatar via 3D-Aware Expression Distillation
par: Jiang, Kaiwen, et autres
Publié: (2025)

PIXART-δ: Fast and Controllable Image Generation with Latent Consistency Models
par: Chen, Junsong, et autres
Publié: (2024)

HAT: Hybrid Attention Transformer for Image Restoration
par: Chen, Xiangyu, et autres
Publié: (2023)

X-Portrait: Expressive Portrait Animation with Hierarchical Motion Attention
par: Xie, You, et autres
Publié: (2024)