Cover Image

FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Fan, Ziyang, Chen, Keyu, Xing, Ruilong, Li, Yulin, Jiang, Li, Tian, Zhuotao
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior
by: Li, Yulin, et al.
Published: (2025)

Token Merging via Spatiotemporal Information Mining for Surgical Video Understanding
by: Jiang, Xixi, et al.
Published: (2025)

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

XS-VID: An Extremely Small Video Object Detection Dataset
by: Guo, Jiahao, et al.
Published: (2024)

Video Token Merging for Long-form Video Understanding
by: Lee, Seon-Ho, et al.
Published: (2024)

Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models
by: Li, Jinlong, et al.
Published: (2026)

Generalized Decoupled Learning for Enhancing Open-Vocabulary Dense Perception
by: Wang, Junjie, et al.
Published: (2025)

FastVID: Dynamic Density Pruning for Fast Video Large Language Models
by: Shen, Leqi, et al.
Published: (2025)

Fast-Slow Efficient Training for Multimodal Large Language Models via Visual Token Pruning
by: Zhang, Dingkun, et al.
Published: (2026)

STEAR: Layer-Aware Spatiotemporal Evidence Intervention for Hallucination Mitigation in Video Large Language Models
by: Fan, Linfeng, et al.
Published: (2026)

Less Languages, Less Tokens: An Efficient Unified Logic Cross-lingual Chain-of-Thought Reasoning Framework
by: Zhang, Chenyuan, et al.
Published: (2026)

Unified Language-driven Zero-shot Domain Adaptation
by: Yang, Senqiao, et al.
Published: (2024)

VideoMerge: Towards Training-free Long Video Generation
by: Zhang, Siyang, et al.
Published: (2025)

Enhancing Spatial Reasoning in Multimodal Large Language Models through Reasoning-based Segmentation
by: Ning, Zhenhua, et al.
Published: (2025)

HoliTom: Holistic Token Merging for Fast Video Large Language Models
by: Shao, Kele, et al.
Published: (2025)

Mitigating Object Hallucinations via Sentence-Level Early Intervention
by: Peng, Shangpin, et al.
Published: (2025)

SpatialVID: A Large-Scale Video Dataset with Spatial Annotations
by: Wang, Jiahao, et al.
Published: (2025)

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

LISA: Reasoning Segmentation via Large Language Model
by: Lai, Xin, et al.
Published: (2023)

Importance-Based Token Merging for Efficient Image and Video Generation
by: Wu, Haoyu, et al.
Published: (2024)

TokenDial: Continuous Attribute Control in Text-to-Video via Spatiotemporal Token Offsets
by: Liu, Zhixuan, et al.
Published: (2026)

Efficient Reasoning with Balanced Thinking
by: Li, Yulin, et al.
Published: (2026)

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

CI-VID: A Coherent Interleaved Text-Video Dataset
by: Ju, Yiming, et al.
Published: (2025)

TempMe: Video Temporal Token Merging for Efficient Text-Video Retrieval
by: Shen, Leqi, et al.
Published: (2024)

ReToMe-VA: Recursive Token Merging for Video Diffusion-based Unrestricted Adversarial Attack
by: Gao, Ziyi, et al.
Published: (2024)

A General and Efficient Training for Transformer via Token Expansion
by: Huang, Wenxuan, et al.
Published: (2024)

Benchmarking and Evolving Reason-Reflect-Rectify for Reflective Visual Generation
by: Wang, Junjie, et al.
Published: (2026)

Vision Token Reduction via Attention-Driven Self-Compression for Efficient Multimodal Large Language Models
by: Deniz, Omer Faruk, et al.
Published: (2026)

Beyond Training: Dynamic Token Merging for Zero-Shot Video Understanding
by: Zhang, Yiming, et al.
Published: (2024)

Unlocking Full Efficiency of Token Filtering in Large Language Model Training
by: Chai, Di, et al.
Published: (2025)

FlashVideo: Flowing Fidelity to Detail for Efficient High-Resolution Video Generation
by: Zhang, Shilong, et al.
Published: (2025)

Compressing Sequences in the Latent Embedding Space: $K$-Token Merging for Large Language Models
by: Xu, Zihao, et al.
Published: (2026)

On the Limits of Token Reduction for Efficient Unified Vision Language Training
by: Chen, Siyi, et al.
Published: (2026)

FastSTAR: Spatiotemporal Token Pruning for Efficient Autoregressive Video Synthesis
by: Yune, Sungwoong, et al.
Published: (2026)

R2E-VID: Two-Stage Robust Routing via Temporal Gating for Elastic Edge-Cloud Video Inference
by: Yang, Zheming, et al.
Published: (2026)

KeyVID: Keyframe-Aware Video Diffusion for Audio-Synchronized Visual Animation
by: Wang, Xingrui, et al.
Published: (2025)

EvoMerge: Neuroevolution for Large Language Models
by: Jiang, Yushu
Published: (2024)

Video, How Do Your Tokens Merge?
by: Pollard, Sam, et al.
Published: (2025)

ClimateVID -- Social Media Videos Analysis and Challenges Involved
by: Xu, Shiqi, et al.
Published: (2026)