About me

I am a second-year PhD student at the Institute for AI Industry Research (AIR), Tsinghua University, advised by Prof. Jingjing Liu (aka JJ) and Prof. Hao Zhou.

My research interest lies in self-supervised learning, multimodal large foundation models.

I love coding and playing with complicated source code, love the same feeling as Greg Brockman:

that feeling when you finally understand a piece of previously inscrutable code

I am interested in physics and in approximating the truth of our universe.

Research Persuit

I am devoting all myself to pursue the birth of safe super intelligence. Only after this goal is achieved can I truly retire :).

Long way to go.

Dec, 23: We are thrilled to announce Emu2, the most capable open-weight vision-language multimodal model, which can both understand and draw images. Try the demo!

2023.01 – Now, Beijing Academy of Artificial Intelligence (BAAI), mentored by Xinlong Wang, Quan Sun and Yue Cao.

* Equal Contribution

Qiying Yu*, Quan Sun*, Xiaosong Zhang, Yufeng Cui, Fan Zhang, Yue Cao, Xinlong Wang, Jingjing Liu. CapsFusion: Rethinking Image-text Data at Scale. CVPR 2024. [code&data]
(Featured in HuggingFace🤗 Daily Paper)

Quan Sun*, Yufeng Cui*, Xiaosong Zhang*, Fan Zhang*, Qiying Yu*, Zhengxiong Luo, Yueze Wang, Yongming Rao, Jingjing Liu, Tiejun Huang, Xinlong Wang. Generative Multimodal Models are In-Context Learners. CVPR 2024. [code] [project] [demo]
(Featured in HuggingFace🤗 Daily Paper)

Quan Sun*, Qiying Yu*, Yufeng Cui*, Fan Zhang*, Xiaosong Zhang*, Yueze Wang, Hongcheng Gao, Jingjing Liu, Tiejun Huang, Xinlong Wang. Generative Pretraining in Multimodality. ICLR 2024. [code]
(Featured in HuggingFace🤗 Daily Paper)

Qiying Yu, Yudi Zhang, Yuyan Ni, Shikun Feng, Yanyan Lan, Hao Zhou, Jingjing Liu. Multimodal Molecular Pretraining via Modality Blending. ICLR 2024.

Qiying Yu, Jieming Lou, Xianyuan Zhan, Qizhang Li, Wangmeng Zuo, Yang Liu and Jingjing Liu. Adversarial Contrastive Learning via Asymmetric InfoNCE. ECCV 2022. [code]

Quan Sun*, Jinsheng Wang*, Qiying Yu*, Yufeng Cui, Fan Zhang, Xiaosong Zhang, Xinlong Wang. EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters. [code]
(Featured in HuggingFace🤗 Daily Paper)

Reviewer: NIPS 2023, ICLR 2023, CVPR 2024, ICML 2024

Below are some persons from around the world whom I greatly admire:

Andrej Karpathy, Mu Li: They have had a profound impact on me with their incredible open-source spirit.

Ilya Sutskever, Sam Altman

Greg Brockman: President of OpenAI. It is amazing that he still writes low-level code at such a high-level position.

Edward Witten: He revolutionized my understanding of the physical world surrounding me.