Imagen de Portada

EDIT: Enhancing Vision Transformers by Mitigating Attention Sink through an Encoder-Decoder Architecture

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Feng, Wenfeng, Wang, Hongxiang, Wang, Jianlong, Zhang, Xin, Zhao, Jingjing, Liang, Yueyue, Chen, Xiang, Han, Duokui
Formato:	Preprint
Publicado:	2025
Materias:	Computer Vision and Pattern Recognition Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Artifacts and Attention Sinks: Structured Approximations for Efficient Vision Transformers
por: Lu, Andrew, et al.
Publicado: (2025)

Mitigating Hallucination in Large Vision-Language Models through Aligning Attention Distribution to Information Flow
por: Zhao, Jianfei, et al.
Publicado: (2025)

SAGE: Sink-Aware Grounded Decoding for Multimodal Hallucination Mitigation
por: Shukla, Tripti, et al.
Publicado: (2026)

On the Nature of Attention Sink that Shapes Decoding Strategy in Omni-LLMs
por: Yoo, Suho, et al.
Publicado: (2026)

Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention
por: Zhao, Jianfei, et al.
Publicado: (2025)

VEGAS: Mitigating Hallucinations in Large Vision-Language Models via Vision-Encoder Attention Guided Adaptive Steering
por: Wang, Zihu, et al.
Publicado: (2025)

Enhancing Temporal Understanding in Video-LLMs through Stacked Temporal Attention in Vision Encoders
por: Rasekh, Ali, et al.
Publicado: (2025)

Enhanced Encoder-Decoder Architecture for Accurate Monocular Depth Estimation
por: Das, Dabbrata, et al.
Publicado: (2024)

SinkTrack: Attention Sink based Context Anchoring for Large Language Models
por: Liu, Xu, et al.
Publicado: (2026)

When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language Models
por: Choi, Jiho, et al.
Publicado: (2026)

An Enhanced Encoder-Decoder Network Architecture for Reducing Information Loss in Image Semantic Segmentation
por: Gao, Zijun, et al.
Publicado: (2024)

Attention Sinks in Diffusion Transformers: A Causal Analysis
por: Wu, Fangzheng, et al.
Publicado: (2026)

Vision Transformers are Circulant Attention Learners
por: Han, Dongchen, et al.
Publicado: (2025)

S2AFormer: Strip Self-Attention for Efficient Vision Transformer
por: Xu, Guoan, et al.
Publicado: (2025)

Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding
por: Wang, Xintong, et al.
Publicado: (2024)

GPT-IMAGE-EDIT-1.5M: A Million-Scale, GPT-Generated Image Dataset
por: Wang, Yuhan, et al.
Publicado: (2025)

DarSwin-Unet: Distortion Aware Encoder-Decoder Architecture
por: Athwale, Akshaya, et al.
Publicado: (2024)

Beyond the Vision Encoder: Identifying and Mitigating Spatial Bias in Large Vision-Language Models
por: Zhu, Yingjie, et al.
Publicado: (2025)

To Sink or Not to Sink: Visual Information Pathways in Large Vision-Language Models
por: Luo, Jiayun, et al.
Publicado: (2025)

Mitigating Attention Sinks and Massive Activations in Audio-Visual Speech Recognition with LLMs
por: Anand, et al.
Publicado: (2025)

AnchorFormer: Differentiable Anchor Attention for Efficient Vision Transformer
por: Shan, Jiquan, et al.
Publicado: (2025)

FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures
por: Lu, Hao, et al.
Publicado: (2024)

Learning A Physical-aware Diffusion Model Based on Transformer for Underwater Image Enhancement
por: Zhao, Chen, et al.
Publicado: (2024)

Attention Based Encoder Decoder Model for Video Captioning in Nepali (2023)
por: Parajuli, Kabita, et al.
Publicado: (2023)

Mixture of Decoding: An Attention-Inspired Adaptive Decoding Strategy to Mitigate Hallucinations in Large Vision-Language Models
por: Chen, Xinlong, et al.
Publicado: (2025)

Representative Attention For Vision Transformers
por: Li, Yuntong, et al.
Publicado: (2026)

YARD: Y-Architecture Register Decoding for Efficient Hallucination Mitigation in Large Vision-Language Models
por: Chen, Ting, et al.
Publicado: (2026)

MED-VT++: Unifying Multimodal Learning with a Multiscale Encoder-Decoder Video Transformer
por: Karim, Rezaul, et al.
Publicado: (2023)

Buster: Implanting Semantic Backdoor into Text Encoder to Mitigate NSFW Content Generation
por: Zhao, Xin, et al.
Publicado: (2024)

Don't Deceive Me: Mitigating Gaslighting through Attention Reallocation in LMMs
por: Jiao, Pengkun, et al.
Publicado: (2025)

IKOD: Mitigating Visual Attention Degradation in Large Vision-Language Models
por: Yang, Jiabing, et al.
Publicado: (2025)

Spiking Vision Transformer with Saccadic Attention
por: Wang, Shuai, et al.
Publicado: (2025)

PA-Attack: Guiding Gray-Box Attacks on LVLM Vision Encoders with Prototypes and Attention
por: Mei, Hefei, et al.
Publicado: (2026)

GLID: Pre-training a Generalist Encoder-Decoder Vision Model
por: Liu, Jihao, et al.
Publicado: (2024)

Spotlight and Shadow: Attention-Guided Dual-Anchor Introspective Decoding for MLLM Hallucination Mitigation
por: Wu, Yebo, et al.
Publicado: (2026)

BSViT: A Burst Spiking Vision Transformer for Expressive and Efficient Visual Representation Learning
por: Peng, Hongxiang, et al.
Publicado: (2026)

TERDNet: Transformer Encoder-Recurrent Decoder Network for Scene Change Detection
por: Yoon, Jiae, et al.
Publicado: (2026)

Adapting LLaMA Decoder to Vision Transformer
por: Wang, Jiahao, et al.
Publicado: (2024)

DASViT: Differentiable Architecture Search for Vision Transformer
por: Wu, Pengjin, et al.
Publicado: (2025)

EagleVision: Object-level Attribute Multimodal LLM for Remote Sensing
por: Jiang, Hongxiang, et al.
Publicado: (2025)