Visual Token Compression

Follow this topic Ask about this topic

Momentum

20 papers in the last four weeks, up 67% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 184

All topics

Spatial-Aware Reduction Framework: Towards Efficient and Faithful Visual State Space Models
Jun 18, 2026Jindi Lv, Aoyu Li, Yuhao Zhou +6Visual SSMs Visual Token Pruning
Sichuan University · GigaAI · Tsinghua University
FrequencyFormer: A Co-Designed Sensor-to-Processor Pipeline for Frequency-Domain Vision Transformer Inference
Jun 17, 2026Chengwei Zhou, Ovishake Sen, Xuming Chen +5Edge Inference Efficient ViTs
Case Western Reserve University, USA · University of Florida, USA · New Jersey Institute of Technology, USA
RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer
Jun 16, 2026Jinhao You, Shuo Lyu, Zhuohang Lyu +53D ViTs Multi-View 3D Reconstruction
University of Pennsylvania · University of California, Irvine · Nanyang Technological University +1
TivTok: Broadcasting Time-Invariant Tokens for Scalable Video Tokenization
Jun 16, 2026Weiliang Chen, Yuanhui Huang, Xuebo Wang +1Visual Tokenization Video Token Compression
Department of Electronic Engineering, Tsinghua University, Beijing, 100084, China · Department of Automation, Tsinghua University, Beijing, 100084, China · Kuaishou Technology, Beijing, China
DCP-Prune: Ultra-Low Token Pruning with Distribution Consistency Preservation
Jun 15, 2026Xifeng Xue, Xiaokang Wang, Zirui Li +2Efficient VLM Inference Visual Token Pruning
College of Computer Science, Nankai University, Tianjin, China · Nanjing University of Posts and Telecommunications, Nanjing, China
One Layer's Trash is Another Layer's Treasure: Adaptive Layer-wise Visual Token Selection in LVLMs
Jun 12, 2026Yongru Chen, Kai Zhang, Zeliang Zong +4Efficient VLM Inference Large Vision-Language Models
Hikvision Research Institute · East China Normal University · Peking University
HiLo-Token: Input-Adaptive High-Low Frequency Token Compression for Efficient Image Editing
Jun 11, 2026Haoran You, Yotam Nitzan, Lingzhi Zhang +7Diffusion Transformer Diffusion-Based Image Editing
Adobe ART AI Lab · Adobe Research
HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers
Jun 11, 2026Guozhen Zhang, Xuerui Qiu, Yutao Cui +11Unified Multimodal Models Vision Transformer
Nanjing University · Work done during internship at Tencent Hunyuan. · CASIA +3
Q-Fold: Query-Aware Focus-Context Spatio-Temporal Folding for Long Video Understanding
Jun 10, 2026Biao Tang, Xu Chen, Shuxiang Gou +3Multimodal Large Language Models Long-Video Understanding
Shenzhen Campus of Sun Yat-sen University · Harbin Institute of Technology, Shenzhen · Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China
Last But Not Least: Boundary Attention CalibratiON for Multimodal KV Cache Compression
Jun 10, 2026Tianhao Chen, Yuheng Wu, Kelu Yao +3KV Caching Multimodal Large Language Models
KAIST · Zhejiang Laboratory · The Chinese University of Hong Kong +1
IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder
Jun 9, 2026Yitong Chen, Zijie Diao, Junke Wang +5Visual Representation Learning Image Tokenization
Institute of Trustworthy Embodied AI, Fudan University · Shanghai Innovation Institute · University of Maryland, College Park
RAPID: Layer-Wise Redundancy-Aware Pruning and Importance-Driven Token Merging for Efficient ViT
Jun 6, 2026Kyumin Choi, Ikbeom JangVision Transformer Efficient ViTs
Hankuk University of Foreign Studies Seoul, South Korea
AdaTok: Self-Budgeting Image Tokenization with Quality-Preserving Dynamic Tokens
Jun 5, 2026Xiaocheng Lu, Yuxi Chen, Jie Zhang +5Image Tokenization Visual Tokenization
The Hong Kong University of Science and Technology · The Hong Kong Polytechnic University
Stream3D-VLM: Online 3D Spatial Understanding with Incremental Geometry Priors
Jun 5, 2026Hanxun Yu, Xuan Qu, Lei Ke +43D Spatial Reasoning 3D VQA
1Zhejiang University · 2Tencent Hunyuan · 3HKUST +1
Adaptive Tokenisation Via Temporal Redundancy Masking And Latent Inpainting
Jun 4, 2026Kevin Dave, Sai Aditya Patkuri, Chhaya Kumar Das +3Visual Tokenization Video Token Compression
Phronetic AI · IISc Bangalore · IIT Bombay
Balancing Image Compression and Generation with Bootstrapped Tokenization
Jun 4, 2026Haozhe Chi, Jinghan Li, Hao Jiang +4Learned Image Compression Image Tokenization
1Peking University · 2Central Media Technology Institute, Huawei
Query-based Cross-Modal Projector Bolstering Mamba Multimodal LLM
Jun 3, 2026SooHwan Eom, Jay Shim, Gwanhyeong Koo +4Mamba Efficient VLM Inference
Korea Advanced Institute of Science and Technology / Korea, Republic of · University of Illinois in Urbana-Champaign / United States of America · Korea University / Korea, Republic of
ChannelTok: Efficient Flexible-Length Vision Tokenization
Jun 3, 2026Sukriti Paul, Arpit Bansal, Tom GoldsteinImage Tokenization Visual Tokenization
University of Maryland, College Park
Zero-Shot 3D Question Answering via Hierarchical View-to-Token Transportation
Jun 2, 2026Dongsheng Wang, Dawei Su, Hui Huang3D VQA Visual Token Compression
College of Computer Science and Software Engineering, Shenzhen University, China.
AdaCodec: A Predictive Visual Code for Video MLLMs
Jun 1, 2026Haowen Hou, Zhen Huang, Zheming Liang +8Predictive Coding Efficient VLM Inference
Shanghai Jiao Tong University · Shanghai Innovation Institute
InfoMerge: Information-aware Token Compression for Efficient Video Large Language Models
Jun 1, 2026Xinxin Liu, Shiwei Gan, Xiao Liu +3Efficient VLM Inference Video-Language Models
State Key Laboratory of Novel Software Technology, Nanjing University Nanjing 210023, China
EvoCut: Multi-Layer Evolution-Aware Visual Token Compression for Efficient Large Vision-Language Models
Jun 1, 2026Hongyu Lu, Feng Zhang, Wenwei Jin +5Efficient VLM Inference Large Vision-Language Models
Harbin Institute of Technology · Xiaohongshu · Fudan University
Improving Visual Token Reduction via Rectifying Distortions for Efficient Multimodal LLM Inference
Jun 1, 2026Hyeonwoo Cho, Donghyeon Baek, Yewon Kim +1Efficient Multimodal Inference Multimodal Large Language Models
Yonsei University · Korea Institute of Science and Technology (KIST)
Wavelet as Tokenizer: Preliminary Results on a Shared Wavelet Token Schema for Natural Signals
May 30, 2026Shenghao DingWavelet Methods Multimodal Token Compression
Yet Another AI
ETC: Extreme Token Compression via Task-aware Visual Information Distillation in VLMs
May 30, 2026Yiling Gao, Hongchen Wei, Zhenzhong ChenEfficient VLM Inference Large Vision-Language Models
School of Remote Sensing and Information Engineering, Wuhan University
PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding
May 28, 2026Selim Kuzucu, Alessio Tonioni, Vasile Lup +3Image Tokenization Efficient VLM Inference
Max Planck Institute for Informatics, SIC · Google · Technical University of Munich
EarlyTom: Early Token Compression Completes Fast Video Understanding
May 28, 2026Hesong Wang, Xin Jin, Lu Lu +4Video Understanding Efficient VLM Inference
Zhejiang University · Westlake University · Alibaba Cloud Computing
OccamToken: Efficient VLM Inference with Training-Free and Budget-Adaptive Token Pruning
May 28, 2026Geng Li, Guohao Chen, Ting Chen +6Efficient VLM Inference Large Vision-Language Models
1Nanyang Technological University (NTU)
CIVIC: End-to-End Sequence Compactness for Efficient Vision-Language Models
May 27, 2026Fengze Yang, Bo Yu, Xuewen Luo +2Efficient VLM Inference KV-Cache Compression
Department of Civil & Environmental Engineering University of Utah 201 Presidents’ Cir, Salt Lake City, UT 84112, USA
Structure over Pixels: Learning Variable-Length Visual Programs
May 26, 2026Piotr Wyrwiński, Kacper Dobek, Krzysztof KrawiecVisual Representation Learning Image Tokenization
Institute of Computing Science Poznan University of Technology, Poznan, Poland

Jun 18, 2026, cs.CVJ/K move · Enter open · S save

Spatial-Aware Reduction Framework: Towards Efficient and Faithful Visual State Space Models

Jindi Lv, Aoyu Li, Yuhao Zhou, Zheng Zhu+5

Sichuan University · GigaAI · Tsinghua University

Visual SSMs Visual Token Pruning Visual Token Compression

Open paper AskPDF arXiv

Mamba demonstrates strong efficiency in modeling long visual sequences. However, when token reduction is applied to structurally enhanced Mamba variants, these models exhibit a severe performance collapse. We attribute this degradation to the spatially agnostic nature of existing reduction methods, which violate the two-dimensional structural premise required by the selective scanning mechanism. In this work, we propose STORM, a spatial-aware token reduction framework designed to maintain structural integrity throughout the compression process. STORM reformulates reduction into a structured operation on spatial units, enforcing localized constraints to maintain both grid topology and neighborhood coherence. As a plug-and-play module, STORM equips existing reduction pipelines with explicit spatial awareness without any training. Empirical results demonstrate that STORM achieves state-of-the-art pruning accuracy across diverse vision Mamba backbones under training-free settings. Notably, STORM delivers a substantial accuracy recovery on VMamba, outperforming prior methods by up to 63.3% in top-1 accuracy. Meanwhile, STORM incurs only a 1.0% accuracy drop on PlainMamba, achieving performance comparable to ViT.

Momentum

Related topics

Latest papers 184