0

Kunchang Li

Papers
20

Cite

Notes

Only stored in your browser.

Attribution

Affiliations & profile
Semantic Scholar
Attribution policy →
20papers

Authored papers

20

Seed1.5-VL Technical Report

arXiv 2025

2025

Make Your Training Flexible: Towards Deployment-Efficient Video Models

ICCV 2025

2025

Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks

arXiv 2024

2024

TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning

arXiv 2024

2024

VideoMamba: State Space Model for Efficient Video Understanding

arXiv 2024

2024

Video Mamba Suite: State Space Model as a Versatile Alternative for Video Understanding

arXiv 2024

2024

Causal Diffusion Transformers for Generative Modeling

arXiv 2024

2024

VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling

arXiv 2024

2024

Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment

CVPR 2025 1

2024

Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel

arXiv 2024

2024

Vlogger: Make Your Dream A Vlog

CVPR 2024 1

2024

TransAgent: Transfer Vision-Language Foundation Models with Heterogeneous Agent Collaboration

arXiv 2024

2024

MUSES: 3D-Controllable Image Generation via Multi-Modal Agent Collaboration

arXiv 2024

2024

VideoEval: Comprehensive Benchmark Suite for Low-Cost Evaluation of Video Foundation Model

arXiv 2024

2024

MVBench: A Comprehensive Multi-modal Video Understanding Benchmark

CVPR 2024 1

2023

Unmasked Teacher: Towards Training-Efficient Video Foundation Models

ICCV 2023 1

2023

You Only Need 90K Parameters to Adapt Light: A Light Weight Transformer for Image Enhancement and Exposure Correction

arXiv 2022

2022

UniFormer: Unified Transformer for Efficient Spatiotemporal Representation Learning

arXiv 2022

2022

UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormer

arXiv 2022

2022

Self-slimmed Vision Transformer

arXiv 2021

2021

Affiliations

No known affiliations.

Frequent co-authors

10

from 20 papers