Haoqing Wang

I am now a PhD candidate (expected to graduate in June 2024) at School of Intelligence Science and Technology, Peking University. I am a member of National Key Lab of General AI, and advised by Professor Zhi-Hong Deng. Previously, I received my Bachelor of Science in mathematics from Beihang University.

Email / Google Scholar / Github

Research

I'm mainly interested in multi/single-modal self-supervised representation learning (i.e., large model pre-training) and its downstream tasks in computer vision, i.e., few-shot learning, detection and segmentation.

	Focus Your Attention when Few-Shot Classification Haoqing Wang, Shibo Jie, Zhi-Hong Deng NeurIPS, 2023. CCF-A. paper / code
	Revisiting the Parameter Efficiency of Adapters from the Perspective of Precision Redundancy Shibo Jie, Haoqing Wang, Zhi-Hong Deng ICCV, 2023. CCF-A. paper / code
	Masked Image Modeling with Local Multi-Scale Reconstruction Haoqing Wang, Yehui Tang, Yunhe Wang, Jianyuan Guo, Zhi-Hong Deng, Kai Han CVPR, 2023 (Highlight Presentation, Top 2.6%). CCF-A. paper / code / blog
	Towards well-generalizing meta-learning via adversarial task augmentation Haoqing Wang, Huiyu Mai, Yuhang Gong, Zhi-Hong Deng Artificial Intelligence, 103875, 2023. CCF-A, IF=14.05. paper / code
	Contrastive Prototypical Network with Wasserstein Confidence Penalty Haoqing Wang, Zhi-Hong Deng ECCV, 2022. CCF-B. paper / code
	Rethinking minimal sufficient representation in contrastive learning Haoqing Wang, Xun Guo, Zhi-Hong Deng, Yan Lu CVPR, 2022 (Oral Presentation, Top 4.2%). CCF-A. paper / code
	Cross-domain few-shot classification via adversarial task augmentation Haoqing Wang, Zhi-Hong Deng IJCAI, 2021. CCF-A. paper / code / blog
	Distributed representations of diseases based on co-occurrence relationship Haoqing Wang, Huiyu Mai, Zhi-Hong Deng, Chao Yang, Luxia Zhang, Huai-yu Wang, Expert Systems with Applications 183, 115418, 2021. CCF-C, IF=8.665. paper
	Few-shot learning with LSSVM base learner and transductive modules Haoqing Wang, Zhi-Hong Deng arXiv preprint arXiv:2009.05786 paper / code
	Fast structured decoding for sequence models Zhiqing Sun, Zhuohan Li, Haoqing Wang, Zi Lin, Di He, Zhi-Hong Deng NeurIPS, 2019. CCF-A. paper / code

Internships
	Vision Intelligence Lab 2023/06/01-2024/02/19, advised by Kang Zhao During the internship, I conduct deep research on video generation with diffusion models, especially on talking head generation.
	Algorithm Application Department 2022/06/15-2022/03/16, advised by Kai Han During the internship, I conduct deep research on masked image modeling and propose a new pretext task, local multi-scale reconstruction, to accelerate representation learning. This work has been accepted by CVPR 2023 as Highlight/Oral presentation (Top 2.5%).
	Multimedia Search and Mining Group 2021/07/20-2022/04/11, advised by Xun Guo During the internship, I conduct deep research on contrastive learning, revealing its shortcomings from a theoretical perspective and proposing solutions. This work has been accepted by CVPR 2022 as Oral presentation (Top 4.2%).

Professional Services

1. Reviewer for CVPR, ICCV, NeurIPS, ICLR, ICML, ECCV, IJCV, TMM, ...

2. Editor of "Introduction to Health Data Science", responsible for "Chapter 9. Machine Learning".