Xianzhi Du

Dalian Medical University — China

Computer Science · Physical Sciences

14h-index1.2kcitations48works6.02yr avg
OpenAlex↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Advanced Neural Network Applications(16), Domain Adaptation and Few-Shot Learning(13), Advanced Image and Video Retrieval Techniques(11), Multimodal Machine Learning Applications(11), Natural Language Processing Techniques(4)

Publications47 total

M2 macrophages modulate the differentiation of CD8 + CD101-TIM3 + T cells via the SPP1‒CD44 pathway, influencing the immunotherapeutic response in NSCLC
Journal of Translational Medicine·2026Open Access
MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer
arXiv (Cornell University)·2025Open Access
AXLearn: Modular, Hardware-Agnostic Large Model Training
ArXiv.org·2025Open Access
Finding Fantastic Experts in MoEs: A Unified Study for Expert Dropping Strategies and Observations
arXiv (Cornell University)·2025Open Access
CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling
arXiv (Cornell University)·2025Open Access
CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling
2025· 1 citedOpen Access
MM1: Methods, Analysis and Insights from Multimodal LLM Pre-training
Lecture notes in computer science·2024· 40 cited
VeCLIP: Improving CLIP Training via Visual-Enriched Captions
Lecture notes in computer science·2024· 8 cited
Apple Intelligence Foundation Language Models
arXiv (Cornell University)·2024· 4 citedOpen Access
Revisiting MoE and Dense Speed-Accuracy Comparisons for LLM Training
arXiv (Cornell University)·2024· 1 citedOpen Access
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
arXiv (Cornell University)·2024· 11 citedOpen Access
Empowering Unsupervised Domain Adaptation with Large-scale Pre-trained Vision-Language Models
2024· 13 cited
VeCLIP: Improving CLIP Training via Visual-enriched Captions
arXiv (Cornell University)·2023· 4 citedOpen Access
Ferret: Refer and Ground Anything Anywhere at Any Granularity
arXiv (Cornell University)·2023· 43 citedOpen Access
Compressing LLMs: The Truth is Rarely Pure and Never Simple
arXiv (Cornell University)·2023· 2 citedOpen Access
AdaMV-MoE: Adaptive Multi-Task Vision Mixture-of-Experts
2023· 47 cited
Guiding Instruction-based Image Editing via Multimodal Large Language Models
arXiv (Cornell University)·2023· 10 citedOpen Access
Mobile V-MoEs: Scaling Down Vision Transformers via Sparse Mixture-of-Experts
arXiv (Cornell University)·2023· 3 citedOpen Access
MOFI: Learning Image Representations from Noisy Entity Annotated Images
arXiv (Cornell University)·2023· 3 citedOpen Access
Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness
arXiv (Cornell University)·2023· 2 citedOpen Access
page 1 of 3Next →

Frequent Co-authors

Tsung-Yi Lin(9), Zhe Gan(9), Yinfei Yang(8), Ruoming Pang(8), Mostafa El‐Khamy(7), Yin Cui(6), Tom Gunter(6), Jungwon Lee(5), Larry S. Davis(5), Xiaodan Song(5), Xiang Kong(5), Nan Du(5), Mark Lee(5), Pengchong Jin(4), Mingxing Tan(4), Quoc V. Le(4), Irwan Bello(4), Barret Zoph(4), Zhangyang Wang(4), Philipp Dufter(4)