Image de couverture de livre

Focus-then-Context: Subject-Centric Progressive Visual Token Reduction for Vision-Language Models

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhao, Yulin, Wang, Yun, Zheng, Dehua, jiang, Borui, Zhang, Zheng
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence 11 F.3.2
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

AGFT: Alignment-Guided Fine-Tuning for Zero-Shot Adversarial Robustness of Vision-Language Models
par: Cui, Yubo, et autres
Publié: (2026)

Towards Lossless Ultimate Vision Token Compression for VLMs
par: Zheng, Dehua, et autres
Publié: (2025)

PPE: Positional Preservation Embedding for Token Compression in Multimodal Large Language Models
par: Huang, Mouxiao, et autres
Publié: (2025)

SimpliPy: A Source-Tracking Notional Machine for Simplified Python
par: Jain, Moida Praneeth, et autres
Publié: (2025)

What if we have 90 minutes only to teach programming?
par: Egri-Nagy, Attila
Publié: (2026)

Double Eisenstein series and modular forms of level $4$
par: Kina, Katsumi
Publié: (2024)

Using Vision + Language Models to Predict Item Difficulty
par: Khan, Samin
Publié: (2026)

Shimura lift of Rankin-Cohen brackets of eigenforms and theta series
par: Wang, Wei
Publié: (2024)

Object-Centric Vision Token Pruning for Vision Language Models
par: Li, Guangyuan, et autres
Publié: (2025)

Weaving Context Across Images: Improving Vision-Language Models through Focus-Centric Visual Chains
par: Zhang, Juntian, et autres
Publié: (2025)

Towards Interpretable Foundation Models for Retinal Fundus Images
par: Mensah, Samuel Ofosu, et autres
Publié: (2026)

On twisted Koecher-Maass series and its integral Kernel
par: Herrera, Fernando
Publié: (2024)

Triple Correlation Sums of Coefficients of Cusp Forms
par: Hulse, Thomas A., et autres
Publié: (2019)

Symmetric multiple Eisenstein series
par: Hara, Takashi, et autres
Publié: (2026)

CLM: Removing the GPU Memory Barrier for 3D Gaussian Splatting
par: Zhao, Hexu, et autres
Publié: (2025)

Inversion by Partial Evaluation: A Reversible Interpreter Experiment
par: Glück, Robert, et autres
Publié: (2024)

Replicate, Reuse, Repeat: Capturing Non-Linear Communication via Session Types and Graded Modal Types
par: Marshall, Danielle, et autres
Publié: (2022)

On obstructions to the Euler system method for Rankin-Selberg convolutions
par: Studnia, Elie
Publié: (2024)

Training-free Token Reduction for Vision Mamba
par: Ma, Qiankun, et autres
Publié: (2025)

Modelling Distributed Applications with Mixed-Choice Stateful Typestates
par: Parrinha, Francisco, et autres
Publié: (2026)

VScan: Rethinking Visual Token Reduction for Efficient Large Vision-Language Models
par: Zhang, Ce, et autres
Publié: (2025)

Current and future roles of artificial intelligence in retinopathy of prematurity
par: Jafarizadeh, Ali, et autres
Publié: (2024)

DRIVE-T: A Methodology for Discriminative and Representative Data Viz Item Selection for Literacy Construct and Assessment
par: Locoro, Angela, et autres
Publié: (2025)

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference
par: Zhang, Yuan, et autres
Publié: (2024)

PVC: Progressive Visual Token Compression for Unified Image and Video Processing in Large Vision-Language Models
par: Yang, Chenyu, et autres
Publié: (2024)

Enhancing Vision-Language Model with Unmasked Token Alignment
par: Liu, Jihao, et autres
Publié: (2024)

HERO: Rethinking Visual Token Early Dropping in High-Resolution Large Vision-Language Models
par: Li, Xu, et autres
Publié: (2025)

Vision-Centric Activation and Coordination for Multimodal Large Language Models
par: Wang, Yunnan, et autres
Publié: (2025)

Seeing the Forest and the Trees: Solving Visual Graph and Tree Based Data Structure Problems using Large Multimodal Models
par: Gutierrez, Sebastian, et autres
Publié: (2024)

Congruences modulo $23$ to $y^2=x^3-23$ are trivial
par: Studnia, Elie
Publié: (2025)

Accelerating Multimodal Large Language Models by Searching Optimal Vision Token Reduction
par: Zhao, Shiyu, et autres
Publié: (2024)

TinyChemVL: Advancing Chemical Vision-Language Models via Efficient Visual Token Reduction and Complex Reaction Tasks
par: Zhao, Xuanle, et autres
Publié: (2025)

Residual paramodularity of a certain Calabi-Yau threefold
par: Dummigan, Neil, et autres
Publié: (2024)

A Vision Centric Remote Sensing Benchmark
par: Adejumo, Abduljaleel, et autres
Publié: (2025)

Compositional Separation of Control Flow and Data Flow
par: Arellanes, Damian
Publié: (2023)

Gradual Guarantee via Step-Indexed Logical Relations in Agda
par: Siek, Jeremy G.
Publié: (2024)

MienCap: Realtime Performance-Based Facial Animation with Live Mood Dynamics
par: Pan, Ye, et autres
Publié: (2025)

Generating real-time detailed ground visualisations from sparse aerial point clouds
par: Murray, Aidan, et autres
Publié: (2025)

Rethinking Token Reduction for Large Vision-Language Models
par: Wang, Yi, et autres
Publié: (2026)

FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression
par: Zhu, Yuke, et autres
Publié: (2024)