Cover Image

DynTok: Dynamic Compression of Visual Tokens for Efficient and Effective Video Understanding

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Hongzhi, Zhang, Jingyuan, Ji, Xingguang, Wang, Qi, Zhang, Fuzheng
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Capybara-OMNI: An Efficient Paradigm for Building Omni-Modal Language Models
by: Ji, Xingguang, et al.
Published: (2025)

Data Metabolism: An Efficient Data Design Schema For Vision Language Model
by: Zhang, Jingyuan, et al.
Published: (2025)

TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic Videos
by: Kong, Fanheng, et al.
Published: (2025)

Clapper: Compact Learning and Video Representation in VLMs
by: Kong, Lingyu, et al.
Published: (2025)

An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes
by: Qi, Ji, et al.
Published: (2025)

Video DataFlywheel: Resolving the Impossible Data Trinity in Video-Language Understanding
by: Wang, Xiao, et al.
Published: (2024)

TokLIP: Marry Visual Tokens to CLIP for Multimodal Comprehension and Generation
by: Lin, Haokun, et al.
Published: (2025)

TrajTok: Learning Trajectory Tokens enables better Video Understanding
by: Zheng, Chenhao, et al.
Published: (2026)

VScan: Rethinking Visual Token Reduction for Efficient Large Vision-Language Models
by: Zhang, Ce, et al.
Published: (2025)

Evaluating Multimodal Large Language Models on Video Captioning via Monte Carlo Tree Search
by: Yu, Linhao, et al.
Published: (2025)

Efficient Whole Slide Pathology VQA via Token Compression
by: Lyu, Weimin, et al.
Published: (2025)

Dynamic Token Compression for Efficient Video Understanding through Reinforcement Learning
by: Wang, Shida, et al.
Published: (2026)

InfoMerge: Information-aware Token Compression for Efficient Video Large Language Models
by: Liu, Xinxin, et al.
Published: (2026)

MergeTok: Unified Continuous and Discrete Visual Tokenization via Token Merging
by: Zhang, Luyuan, et al.
Published: (2026)

HAIC: Improving Human Action Understanding and Generation with Better Captions for Multi-modal Large Language Models
by: Wang, Xiao, et al.
Published: (2025)

DualToken: Towards Unifying Visual Understanding and Generation with Dual Visual Vocabularies
by: Song, Wei, et al.
Published: (2025)

Dyn-Adapter: Towards Disentangled Representation for Efficient Visual Recognition
by: Zhang, Yurong, et al.
Published: (2024)

Video-Text Dataset Construction from Multi-AI Feedback: Promoting Weak-to-Strong Preference Learning for Video Large Language Models
by: Yi, Hao, et al.
Published: (2024)

DINO-Tok: Adapting DINO for Visual Tokenizers
by: Jia, Mingkai, et al.
Published: (2025)

AlignTok: Aligning Visual Foundation Encoders to Tokenizers for Diffusion Models
by: Chen, Bowei, et al.
Published: (2025)

Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization
by: Jin, Yang, et al.
Published: (2024)

UniTok: A Unified Tokenizer for Visual Generation and Understanding
by: Ma, Chuofan, et al.
Published: (2025)

EvoTok: A Unified Image Tokenizer via Residual Latent Evolution for Visual Understanding and Generation
by: Li, Yan, et al.
Published: (2026)

Dynamic-VLM: Simple Dynamic Visual Token Compression for VideoLLM
by: Wang, Han, et al.
Published: (2024)

Principles of Visual Tokens for Efficient Video Understanding
by: Hao, Xinyue, et al.
Published: (2024)

ShortV: Efficient Multimodal Large Language Models by Freezing Visual Tokens in Ineffective Layers
by: Yuan, Qianhao, et al.
Published: (2025)

EgoDyn-Bench: Evaluating Ego-Motion Understanding in Vision-Centric Foundation Models for Autonomous Driving
by: Schäfer, Finn Rasmus, et al.
Published: (2026)

DynImg: Key Frames with Visual Prompts are Good Representation for Multi-Modal Video Understanding
by: Bao, Xiaoyi, et al.
Published: (2025)

Multi-Step Visual Reasoning with Visual Tokens Scaling and Verification
by: Bai, Tianyi, et al.
Published: (2025)

Mobile-Agent-V: A Video-Guided Approach for Effortless and Efficient Operational Knowledge Injection in Mobile Automation
by: Wang, Junyang, et al.
Published: (2025)

DynaTok: Temporally Adaptive and Positional Bias-Aware Token Compression for Video-LLMs
by: Park, Minyoung, et al.
Published: (2026)

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

WeTok: Powerful Discrete Tokenization for High-Fidelity Visual Reconstruction
by: Zhuang, Shaobin, et al.
Published: (2025)

Token Sequence Compression for Efficient Multimodal Computing
by: Omri, Yasmine, et al.
Published: (2025)

WinTok: A Win-Win Hybrid Tokenizer via Decomposing Visual Understanding and Generation with Transferable Tokens
by: Guo, Yiwei, et al.
Published: (2026)

Vision-centric Token Compression in Large Language Model
by: Xing, Ling, et al.
Published: (2025)

MMViR: A Multi-Modal and Multi-Granularity Representation for Long-range Video Understanding
by: Li, Zizhong, et al.
Published: (2026)

DynFrame: Adaptive Reasoning-Driven Multimodal Framework with Dynamic Frame Augmentation for Complex Video Understanding
by: Zhang, Peng, et al.
Published: (2026)

From Pixels to Tokens: Revisiting Object Hallucinations in Large Vision-Language Models
by: Shang, Yuying, et al.
Published: (2024)

AdaTok: Adaptive Token Compression with Object-Aware Representations for Efficient Multimodal LLMs
by: Zhang, Xinliang, et al.
Published: (2025)