Qingyang Zhang 张庆阳

I am a first-year PhD student at Tianjin University, supervised by Prof. Changqing Zhang. Currently, I am at an internship at Tencent AI Lab, co-supervised by Yatao Bian .

I received my Bachelor's degree in School of Computer Science and Technology from Tianjin University (2018-2022). After that, I pursued my Master's in the School of Computer Science and Technology at Tianjin University and transitioned into a Ph.D. candidate through the direct doctoral program in 2024.

Email  /  Google Scholar  /  Github

profile photo
News

[2025-01] One paper accepted by ICLR, thanks to all co-authors!

[2024-05] One paper accepted by NeurIPS, thanks to all co-authors!

[2024-05] We release a survey about fusion of low-quality multi-modal data! [arXiv]

[2024-04] Starting an internship at Tencent AI Lab, superivsed by Yatao Bian.

[2023-04] Two paper accepted by ICML including one Oral paper, thanks to all co-authors.

[2022-06] Graduate from Tianjin University.

Survey
clean-usnob Multimodal Fusion on Low-quality Data: A Comprehensive Survey

Qingyang Zhang, Yake Wei, Zongbo Han, Huazhu Fu, Xi Peng, Cheng Deng, Qinghua Hu, Cai Xu, Jie Wen, Di Hu, Changqing Zhang


arXiv / awesome list

A systematical survey about fusion of low-quality multi-modal data.


Publications(* equal contribution)
clean-usnob The Best of both Worlds: On the Dilemma of Out-of-Distribution Detection

Qingyang Zhang, Qiuxuan Feng, Joey Tianyi Zhou, Yatao Bian, Qinghua Hu and Changqing Zhang
NeurIPS, 2024
arXiv / code

Solve conflicts between OOD detection and generalization for dual-optimal performance.

clean-usnob Provable Dynamic Fusion for Low-quality Multimodal Learning

Qingyang Zhang, Haitao Wu, Changqing Zhang, Qinghua Hu, Huazhu Fu, Joey Tianyi Zhou, Xi Peng

ICML, 2023
arXiv / code

Theory-inspired dynimical fuse strategy for quality-varying modalities in real world.

clean-usnob Calibrating Multimodal Learning

Huan Ma, Qingyang Zhang (co-first author), Changqing Zhang, Bingzhe Wu, Huazhu Fu, Joey Tianyi Zhou, Qinghua Hu

ICML, 2023
arXiv / code

Mitigate the greedy nature of multimodal learning by regularizing the model confidence.

Services

Conference Reviewer: ICLR 2022-2024, NeurIPS 2023-2024, ICML 2024

Awards

National Scholarship (twice, 1%) 2022, 2023



Updated at Dec. 2024
Thanks Jon Barron for this amazing template.