Cover Image

VLCache: Computing 2% Vision Tokens and Reusing 98% for Vision-Language Inference

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Qin, Shengling, Yu, Hao, Wu, Chenxin, Li, Zheng, Cao, Yizhong, Zhuge, Zhengyang, Zhou, Yuxin, Yao, Wentao, Zhang, Yi, Wang, Zhengheng, Bai, Shuai, Zhang, Jianwei, Lin, Junyang
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

ToFe: Lagged Token Freezing and Reusing for Efficient Vision Transformer Inference
by: Zhang, Haoyue, et al.
Published: (2025)

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models
by: Chen, Liang, et al.
Published: (2024)

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)

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

TPC-ViT: Token Propagation Controller for Efficient Vision Transformer
by: Zhu, Wentao
Published: (2024)

VLM4VLA: Revisiting Vision-Language-Models in Vision-Language-Action Models
by: Zhang, Jianke, et al.
Published: (2026)

Vision SmolMamba: Spike-Guided Token Pruning for Energy-Efficient Spiking State-Space Vision Models
by: Bai, Dewei, et al.
Published: (2026)

TriSpec: Ternary Speculative Decoding via Lightweight Proxy Verification
by: Jiang, Haoyun, et al.
Published: (2026)

Revisiting Multimodal Positional Encoding in Vision-Language Models
by: Huang, Jie, et al.
Published: (2025)

Efficient Vision-Language Models by Summarizing Visual Tokens into Compact Registers
by: Wen, Yuxin, et al.
Published: (2024)

RetoVLA: Reusing Register Tokens for Spatial Reasoning in Vision-Language-Action Models
by: Koo, Jiyeon, et al.
Published: (2025)

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

Multi‐Dimensional Synergistic Engineering for Boosting Nanozyme Catalysis
by: Yuechun Li, et al.
Published: (2025)

Token-Level Inference-Time Alignment for Vision-Language Models
by: Chen, Kejia, et al.
Published: (2025)

Vision-Language Model Selection and Reuse for Downstream Adaptation
by: Tan, Hao-Zhe, et al.
Published: (2025)

HCC-3D: Hierarchical Compensatory Compression for 98% 3D Token Reduction in Vision-Language Models
by: Zhang, Liheng, et al.
Published: (2025)

Vote&Mix: Plug-and-Play Token Reduction for Efficient Vision Transformer
by: Peng, Shuai, et al.
Published: (2024)

TReX- Reusing Vision Transformer's Attention for Efficient Xbar-based Computing
by: Moitra, Abhishek, et al.
Published: (2024)

Rethinking Token Reduction for Large Vision-Language Models
by: Wang, Yi, et al.
Published: (2026)

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference
by: Zhang, Yuan, et al.
Published: (2024)

Are Bigger Encoders Always Better in Vision Large Models?
by: Li, Bozhou, et al.
Published: (2024)

LeMeViT: Efficient Vision Transformer with Learnable Meta Tokens for Remote Sensing Image Interpretation
by: Jiang, Wentao, et al.
Published: (2024)

TRIO: Token Reduction via Inference-Objective Guidance for Efficient Vision-Language Models
by: Zhang, Haokui, et al.
Published: (2026)

Context-Aware Token Selection and Packing for Enhanced Vision Transformer
by: Zhang, Tianyi, et al.
Published: (2024)

Inference Compute-Optimal Video Vision Language Models
by: Wang, Peiqi, et al.
Published: (2025)

Source-Free Domain Adaptation with Vision-Language Prior
by: Tang, Song, et al.
Published: (2026)

Vision Transformer with Super Token Sampling
by: Huang, Huaibo, et al.
Published: (2022)

Prioritizing Image-Related Tokens Enhances Vision-Language Pre-Training
by: Chen, Yangyi, et al.
Published: (2025)

BTR: Binary Token Representations for Efficient Retrieval Augmented Language Models
by: Cao, Qingqing, et al.
Published: (2023)

Less Is More -- Until It Breaks: Security Pitfalls of Vision Token Compression in Large Vision-Language Models
by: Zhang, Xiaomei, et al.
Published: (2026)

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

AdaptInfer: Adaptive Token Pruning for Vision-Language Model Inference with Dynamical Text Guidance
by: Zhang, Weichen, et al.
Published: (2025)

Slicing Vision Transformer for Flexible Inference
by: Zhang, Yitian, et al.
Published: (2024)

Vision Transformer Computation and Resilience for Dynamic Inference
by: Sreedhar, Kavya, et al.
Published: (2022)

ETA: Evaluating Then Aligning Safety of Vision Language Models at Inference Time
by: Ding, Yi, et al.
Published: (2024)

Parallel Vision Token Scheduling for Fast and Accurate Multimodal LMMs Inference
by: Zhan, Wengyi, et al.
Published: (2025)

Membership Inference Attacks Against Vision-Language Models
by: Hu, Yuke, et al.
Published: (2025)

Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning
by: Zhao, Zhengyang, et al.
Published: (2026)

Learning Spatial Decay for Vision Transformers
by: Mao, Yuxin, et al.
Published: (2025)

End-to-End Vision Tokenizer Tuning
by: Wang, Wenxuan, et al.
Published: (2025)