Cover Image

xGen-MM-Vid (BLIP-3-Video): You Only Need 32 Tokens to Represent a Video Even in VLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ryoo, Michael S., Zhou, Honglu, Kendre, Shrikant, Qin, Can, Xue, Le, Shu, Manli, Park, Jongwoo, Ranasinghe, Kanchana, Savarese, Silvio, Xu, Ran, Xiong, Caiming, Niebles, Juan Carlos
Format:	Preprint
Published:	2024
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

xGen-MM (BLIP-3): A Family of Open Large Multimodal Models
by: Xue, Le, et al.
Published: (2024)

Strefer: Empowering Video LLMs with Space-Time Referring and Reasoning via Synthetic Instruction Data
by: Zhou, Honglu, et al.
Published: (2025)

xGen-VideoSyn-1: High-fidelity Text-to-Video Synthesis with Compressed Representations
by: Qin, Can, et al.
Published: (2024)

Future Optical Flow Prediction Improves Robot Control & Video Generation
by: Ranasinghe, Kanchana, et al.
Published: (2026)

Language Repository for Long Video Understanding
by: Kahatapitiya, Kumara, et al.
Published: (2024)

xGen-small Technical Report
by: Nijkamp, Erik, et al.
Published: (2025)

Active Video Perception: Iterative Evidence Seeking for Agentic Long Video Understanding
by: Wang, Ziyang, et al.
Published: (2025)

SMILE: A Composite Lexical-Semantic Metric for Question-Answering Evaluation
by: Kendre, Shrikant, et al.
Published: (2025)

Too Many Frames, Not All Useful: Efficient Strategies for Long-Form Video QA
by: Park, Jongwoo, et al.
Published: (2024)

Understanding Long Videos with Multimodal Language Models
by: Ranasinghe, Kanchana, et al.
Published: (2024)

Robotic VLA Benefits from Joint Learning with Motion Image Diffusion
by: Fang, Yu, et al.
Published: (2025)

IVRA: Improving Visual-Token Relations for Robot Action Policy with Training-Free Hint-Based Guidance
by: Park, Jongwoo, et al.
Published: (2026)

BLIP3o-NEXT: Next Frontier of Native Image Generation
by: Chen, Jiuhai, et al.
Published: (2025)

CoPT: Unsupervised Domain Adaptive Segmentation using Domain-Agnostic Text Embeddings
by: Mata, Cristina, et al.
Published: (2025)

Pixel Motion as Universal Representation for Robot Control
by: Ranasinghe, Kanchana, et al.
Published: (2025)

AdaVid: Adaptive Video-Language Pretraining
by: Patel, Chaitanya, et al.
Published: (2025)

X-InstructBLIP: A Framework for aligning X-Modal instruction-aware representations to LLMs and Emergent Cross-modal Reasoning
by: Panagopoulou, Artemis, et al.
Published: (2023)

ViUniT: Visual Unit Tests for More Robust Visual Programming
by: Panagopoulou, Artemis, et al.
Published: (2024)

Linear Scaling Video VLMs for Long Video Understanding
by: Eyzaguirre, Cristobal, et al.
Published: (2026)

BLIP3-KALE: Knowledge Augmented Large-Scale Dense Captions
by: Awadalla, Anas, et al.
Published: (2024)

DyMU: Dynamic Merging and Virtual Unmerging for Efficient VLMs
by: Wang, Zhenhailong, et al.
Published: (2025)

BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset
by: Chen, Jiuhai, et al.
Published: (2025)

Pixel Motion Diffusion is What We Need for Robot Control
by: Nguyen, E-Ro, et al.
Published: (2025)

Contra4: Evaluating Contrastive Cross-Modal Reasoning in Audio, Video, Image, and 3D
by: Panagopoulou, Artemis, et al.
Published: (2025)

PerfCodeGen: Improving Performance of LLM Generated Code with Execution Feedback
by: Peng, Yun, et al.
Published: (2024)

VidTok: A Versatile and Open-Source Video Tokenizer
by: Tang, Anni, et al.
Published: (2024)

Learning to Localize Objects Improves Spatial Reasoning in Visual-LLMs
by: Ranasinghe, Kanchana, et al.
Published: (2024)

Enabling High Data Throughput Reinforcement Learning on GPUs: A Domain Agnostic Framework for Data-Driven Scientific Research
by: Lan, Tian, et al.
Published: (2024)

Unifying Specialized Visual Encoders for Video Language Models
by: Chung, Jihoon, et al.
Published: (2025)

LATTE: Learning to Think with Vision Specialists
by: Ma, Zixian, et al.
Published: (2024)

Stateful Token Reduction for Long-Video Hybrid VLMs
by: Jiang, Jindong, et al.
Published: (2026)

Causal Layering via Conditional Entropy
by: Feigenbaum, Itai, et al.
Published: (2024)

Editing Arbitrary Propositions in LLMs without Subject Labels
by: Feigenbaum, Itai, et al.
Published: (2024)

YOSE: You Only Select Essential Tokens for Efficient DiT-based Video Object Removal
by: Wu, Chenyang, et al.
Published: (2026)

GenVidBench: A 6-Million Benchmark for AI-Generated Video Detection
by: Ni, Zhenliang, et al.
Published: (2025)

PruneVid: Visual Token Pruning for Efficient Video Large Language Models
by: Huang, Xiaohu, et al.
Published: (2024)

HieraVid: Hierarchical Token Pruning for Fast Video Large Language Models
by: Guo, Yansong, et al.
Published: (2026)

Shared Imagination: LLMs Hallucinate Alike
by: Zhou, Yilun, et al.
Published: (2024)

LACE: Latent Visual Representation for Cross-Embodiment Learning
by: Jang, Yoo Sung, et al.
Published: (2026)

Unified Spatio-Temporal Token Scoring for Efficient Video VLMs
by: Zhang, Jianrui, et al.
Published: (2026)