Cover Image

TokenCarve: Information-Preserving Visual Token Compression in Multimodal Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Tan, Xudong, Ye, Peng, Tu, Chongjun, Cao, Jianjian, Yang, Yaoxin, Zhang, Lin, Zhou, Dongzhan, Chen, Tao
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Revisiting Multimodal KV Cache Compression: A Frequency-Domain-Guided Outlier-KV-Aware Approach
by: Yang, Yaoxin, et al.
Published: (2025)

Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Models
by: Tan, Xudong, et al.
Published: (2025)

MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer
by: Cao, Jianjian, et al.
Published: (2024)

Sequential Token Merging: Revisiting Hidden States
by: Wen, Yan, et al.
Published: (2025)

ClipSAM: CLIP and SAM Collaboration for Zero-Shot Anomaly Segmentation
by: Li, Shengze, et al.
Published: (2024)

PPE: Positional Preservation Embedding for Token Compression in Multimodal Large Language Models
by: Huang, Mouxiao, et al.
Published: (2025)

Training-Free Efficient Video Generation via Dynamic Token Carving
by: Zhang, Yuechen, et al.
Published: (2025)

Language-Guided Token Compression with Reinforcement Learning in Large Vision-Language Models
by: Cao, Sihan, et al.
Published: (2026)

Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal Models
by: Peng, Tianfan, et al.
Published: (2025)

A Survey of Token Compression for Efficient Multimodal Large Language Models
by: Shao, Kele, et al.
Published: (2025)

LaCo: Efficient Layer-wise Compression of Visual Tokens for Multimodal Large Language Models
by: Liu, Juntao, et al.
Published: (2025)

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

Attention Reallocation: Towards Zero-cost and Controllable Hallucination Mitigation of MLLMs
by: Tu, Chongjun, et al.
Published: (2025)

Compression Tells Intelligence: Visual Coding, Visual Token Technology, and the Unification
by: Jin, Xin, et al.
Published: (2026)

Local Information Matters: Inference Acceleration For Grounded Conversation Generation Models Through Adaptive Local-Aware Token Pruning
by: Bai, Bizhe, et al.
Published: (2025)

A Glimpse to Compress: Dynamic Visual Token Pruning for Large Vision-Language Models
by: Zeng, Quan-Sheng, et al.
Published: (2025)

TrimTokenator: Towards Adaptive Visual Token Pruning for Large Multimodal Models
by: Zhang, Hao, et al.
Published: (2025)

CurveStream: Boosting Streaming Video Understanding in MLLMs via Curvature-Aware Hierarchical Visual Memory Management
by: Wang, Chao, et al.
Published: (2026)

Efficient Token Compression for Vision Transformer with Spatial Information Preserved
by: Mao, Junzhu, et al.
Published: (2025)

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

Window Token Concatenation for Efficient Visual Large Language Models
by: Li, Yifan, et al.
Published: (2025)

Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings
by: Wu, Qiong, et al.
Published: (2024)

On the Adversarial Robustness of Large Vision-Language Models under Visual Token Compression
by: Zhang, Xinwei, et al.
Published: (2026)

Visual-Guided Key-Token Regularization for Multimodal Large Language Model Unlearning
by: Cai, Chengyi, et al.
Published: (2026)

$Δ$-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers
by: Chen, Pengtao, et al.
Published: (2024)

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

TokenSeg: Efficient 3D Medical Image Segmentation via Hierarchical Visual Token Compression
by: Zeng, Sen, et al.
Published: (2026)

Introducing Visual Perception Token into Multimodal Large Language Model
by: Yu, Runpeng, et al.
Published: (2025)

EntropyPrune: Matrix Entropy Guided Visual Token Pruning for Multimodal Large Language Models
by: Wang, Yahong, et al.
Published: (2026)

TrimTokenator-LC: Towards Adaptive Visual Token Pruning for Large Multimodal Models with Long Contexts
by: Zhang, Hao, et al.
Published: (2025)

Recoverable Compression: A Multimodal Vision Token Recovery Mechanism Guided by Text Information
by: Chen, Yi, et al.
Published: (2024)

EvoComp: Learning Visual Token Compression for Multimodal Large Language Models via Semantic-Guided Evolutionary Labeling
by: Song, Jiafei, et al.
Published: (2026)

DyCoke: Dynamic Compression of Tokens for Fast Video Large Language Models
by: Tao, Keda, et al.
Published: (2024)

TokenCom: Vision-Language Model for Multimodal and Multitask Token Communications
by: Jiang, Feibo, et al.
Published: (2026)

Superpixel Tokenization for Vision Transformers: Preserving Semantic Integrity in Visual Tokens
by: Lew, Jaihyun, et al.
Published: (2024)

Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Models
by: Ye, Weihao, et al.
Published: (2024)

Task-Related Token Compression in Multimodal Large Language Models from an Explainability Perspective
by: Lei, Lei, et al.
Published: (2025)

HAWK: Head Importance-Aware Visual Token Pruning in Multimodal Models
by: Zhu, Qihui, et al.
Published: (2026)

Accelerating Streaming Video Large Language Models via Hierarchical Token Compression
by: Wang, Yiyu, et al.
Published: (2025)

Beyond Next-Token Alignment: Distilling Multimodal Large Language Models via Token Interactions
by: Chen, Lin, et al.
Published: (2026)