Cover Image

What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Fan, Yingqi, Tong, Junlong, Zhao, Anhao, Shen, Xiaoyu
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

$\mathcal{V}isi\mathcal{P}runer$: Decoding Discontinuous Cross-Modal Dynamics for Efficient Multimodal LLMs
by: Fan, Yingqi, et al.
Published: (2025)

HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early Exit
by: Wu, Hao, et al.
Published: (2026)

Do MLLMs Really See It: Reinforcing Visual Attention in Multimodal LLMs
by: Ou, Siqu, et al.
Published: (2026)

Prompt-Aware Adapter: Towards Learning Adaptive Visual Tokens for Multimodal Large Language Models
by: Zhang, Yue, et al.
Published: (2024)

Investigating Redundancy in Multimodal Large Language Models with Multiple Vision Encoders
by: Wang, Yizhou, et al.
Published: (2025)

What Kind of Visual Tokens Do We Need? Training-free Visual Token Pruning for Multi-modal Large Language Models from the Perspective of Graph
by: Jiang, Yutao, et al.
Published: (2025)

How Many Visual Tokens Do Multimodal Language Models Need? Scaling Visual Token Pruning with F^3A
by: Huang, YiJie, et al.
Published: (2026)

Do We Really Need a Large Number of Visual Prompts?
by: Kim, Youngeun, et al.
Published: (2023)

Bridging the Gap Between Sparsity and Redundancy: A Dual-Decoding Framework with Global Context for Map Inference
by: Shen, Yudong, et al.
Published: (2025)

Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference
by: Lin, Zhihang, et al.
Published: (2024)

A Comprehensive Study on Visual Token Redundancy for Discrete Diffusion-based Multimodal Large Language Models
by: Li, Duo, et al.
Published: (2025)

ViCA: Efficient Multimodal LLMs with Vision-Only Cross-Attention
by: Liu, Wenjie, et al.
Published: (2026)

UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encoding
by: Jiao, Yang, et al.
Published: (2025)

Break the Visual Perception: Adversarial Attacks Targeting Encoded Visual Tokens of Large Vision-Language Models
by: Wang, Yubo, et al.
Published: (2024)

Understanding and Harnessing Sparsity in Unified Multimodal Models
by: He, Shwai, et al.
Published: (2025)

Perception Tokens Enhance Visual Reasoning in Multimodal Language Models
by: Bigverdi, Mahtab, et al.
Published: (2024)

StreamingThinker: Large Language Models Can Think While Reading
by: Tong, Junlong, et al.
Published: (2025)

Speak While Watching: Unleashing TRUE Real-Time Video Understanding Capability of Multimodal Large Language Models
by: Lin, Junyan, et al.
Published: (2026)

Enhancing Instruction-Following Capability of Visual-Language Models by Reducing Image Redundancy
by: Yang, Te, et al.
Published: (2024)

LFTR: Learning-Free Token Reduction for Multimodal Large Language Models
by: Zhao, Zihui, et al.
Published: (2025)

GroundVTS: Visual Token Sampling in Multimodal Large Language Models for Video Temporal Grounding
by: Fan, Rong, et al.
Published: (2026)

Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models
by: Ma, Chuofan, et al.
Published: (2024)

UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking
by: Wu, Hao, et al.
Published: (2026)

Do Multimodal Agents Really Benefit from Tool Use? A Systematic Study of Capability Gains
by: Guo, Garvin, et al.
Published: (2026)

GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models
by: Zheng, Shurong, et al.
Published: (2026)

Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices
by: Lin, Junyan, et al.
Published: (2025)

Unveiling the Response of Large Vision-Language Models to Visually Absent Tokens
by: Kim, Sohee, et al.
Published: (2025)

Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models
by: Zhang, Jialiang, et al.
Published: (2026)

AdaTP: Attention-Debiased Token Pruning for Video Large Language Models
by: Sun, Fengyuan, et al.
Published: (2025)

PAR: Prompt-Aware Token Reduction Method for Efficient Large Multimodal Models
by: Liu, Yingen, et al.
Published: (2024)

See What You Are Told: Visual Attention Sink in Large Multimodal Models
by: Kang, Seil, et al.
Published: (2025)

ForestPrune: High-ratio Visual Token Compression for Video Multimodal Large Language Models via Spatial-Temporal Forest Modeling
by: Ju, Shaobo, et al.
Published: (2026)

QAPruner: Quantization-Aware Vision Token Pruning for Multimodal Large Language Models
by: Wang, Xinhao, et al.
Published: (2026)

FCoT-VL:Advancing Text-oriented Large Vision-Language Models with Efficient Visual Token Compression
by: Li, Jianjian, et al.
Published: (2025)

Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens
by: Yang, Zeyuan, et al.
Published: (2025)

Understand, Think, and Answer: Advancing Visual Reasoning with Large Multimodal Models
by: Zhan, Yufei, et al.
Published: (2025)

Do Vision-Language Models Really Understand Visual Language?
by: Hou, Yifan, et al.
Published: (2024)

Context Guided Transformer Entropy Modeling for Video Compression
by: Tong, Junlong, et al.
Published: (2025)

Uncovering Bias in Large Vision-Language Models with Counterfactuals
by: Howard, Phillip, et al.
Published: (2024)

Multimodal Language Models See Better When They Look Shallower
by: Chen, Haoran, et al.
Published: (2025)