Keyu Tian
硕士研究生 at 北京大学
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolKeyu Tian is a Master's student in Computer Science at Peking University with seven years of research and industry experience focused on multimodal learning, video generation, and reinforcement learning. They blend self-supervised and generative-model expertise to tackle open-ended AI challenges that span temporal and interactive domains. Keyu's background includes research internships at SenseTime, a remote collaboration with Oxford's Torr Vision Group, and a multi-year research role at ByteDance AI Lab's VC Team, giving them hands-on experience deploying video and vision models at scale. This profile reflects a practitioner who bridges academic rigor and product-facing research, comfortable collaborating internationally and driving scalable multimodal systems.
7 years of coding experience
Master CS, Master CS at Peking University
Github Skills (57)
sparse10
convolutional-neural-networks10
ssl10
hierarchical10
python10
pre-trained-model10
re-identification10
pretrain10
machine-learning10
gpt-210
voc10
deep-learning10
self-supervised-learning10
gpu10
kitti-dataset10
Programming languages (4)
Github contributions (5)
Ke-Coding/Data-Structure-Cpp11
Mar 2019 - Jul 2019
Contributions:78 commits, 23 pushes, 1 branch in 3 months
keyu-tian/SparK
Jan 2023 - Mar 2023
[ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"
Contributions:2 releases, 16 commits, 109 pushes in 2 months
bertconvnethierarchicalsparsevisual-recognition
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.