Hengyuan Zhao

Henry Hengyuan Zhao (赵恒远)

Hi👋, this is Henry. I'm a third-year PhD student at Show Lab, National University of Singapore, under the supervision of Prof. Mike Zheng Shou. In the past time, I've been fortunate to work with Pan Zhou.

I am generally interested in multimodal understanding and AI automation. My recent focus is on building intelligent AI for solving real-world problems and exploring the future role of current AI model.

1. Define a right problem to work on. 2. Solve it from the first principle.

Links: Email / GitHub / Google Scholar / Twitter

📢 News

[09/2024] One paper is accpeted by NeurIPS 2024.
[07/2024] One paper is accpeted by ECCV 2024.
[09/2023] One paper is accpeted by IJCV 2023.
[08/2023] One paper is accpeted by TPAMI 2023.

🌺 Publications

See full publications here.

	LOVA3: Learning to Visual Question Answering, Asking and Assessment Henry Hengyuan Zhao, Pan Zhou, Difei Gao, Mike Zheng Shou Neural Information Processing Systems (NeurIPS 2024) Can MLLMs have asking and assessment abilities similar to how humans learning knowledges? Paper \| Codes
	Genixer: Empowering Multimodal Large Language Models as a Powerful Data Generator Henry Hengyuan Zhao, Pan Zhou, Mike Zheng Shou European Conference on Computer Vision (ECCV 2024) If MLLMs represent the state of the art, how do they perform in data generation? We developed two data generators for nine vision-language (VL) tasks. Paper \| Codes
	SCT: A Simple Baseline for Parameter-Efficient Fine-Tuning via Salient Channels Henry Hengyuan Zhao, Pichao Wang, Yuyang Zhao, Hao Luo, Fan Wang, Mike Zheng Shou International Journal of Computer Vision (IJCV 2023) We found that tuning only a small number of task-specific channels, referred to as salient channels, is sufficient. This work represents a remarkable reduction of 780x in parameter costs compared to its full fine-tuning counterpart. Paper \| Codes
	Evaluating the Generalization Ability of Super-Resolution Networks Yihao Liu, Hengyuan Zhao, Jinjin Gu, Yu Qiao, Chao Dong IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI 2023) Paper \| Codes
	ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic Xiangtao Kong, Hengyuan Zhao, Yu Qiao, Chao Dong Computer Vision and Pattern Recognition (CVPR 2021) Paper \| Codes
	Efficient Image Super-Resolution Using Pixel Attention Hengyuan Zhao, Xiangtao Kong, Jingwen He, Yu Qiao, Chao Dong European Conference on Computer Vision Workshops (ECCVW 2020) Over 390 citations Paper \| Codes

	LOVA3: Learning to Visual Question Answering, Asking and Assessment Henry Hengyuan Zhao, Pan Zhou, Difei Gao, Mike Zheng Shou Neural Information Processing Systems (NeurIPS 2024) Can MLLMs have asking and assessment abilities similar to how humans learning knowledges? Paper \| Codes
	Genixer: Empowering Multimodal Large Language Models as a Powerful Data Generator Henry Hengyuan Zhao, Pan Zhou, Mike Zheng Shou European Conference on Computer Vision (ECCV 2024) If MLLMs represent the state of the art, how do they perform in data generation? We developed two data generators for nine vision-language (VL) tasks. Paper \| Codes
	SCT: A Simple Baseline for Parameter-Efficient Fine-Tuning via Salient Channels Henry Hengyuan Zhao, Pichao Wang, Yuyang Zhao, Hao Luo, Fan Wang, Mike Zheng Shou International Journal of Computer Vision (IJCV 2023) We found that tuning only a small number of task-specific channels, referred to as salient channels, is sufficient. This work represents a remarkable reduction of 780x in parameter costs compared to its full fine-tuning counterpart. Paper \| Codes
	Evaluating the Generalization Ability of Super-Resolution Networks Yihao Liu, Hengyuan Zhao, Jinjin Gu, Yu Qiao, Chao Dong IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI 2023) Paper \| Codes
	ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic Xiangtao Kong, Hengyuan Zhao, Yu Qiao, Chao Dong Computer Vision and Pattern Recognition (CVPR 2021) Paper \| Codes
	Efficient Image Super-Resolution Using Pixel Attention Hengyuan Zhao, Xiangtao Kong, Jingwen He, Yu Qiao, Chao Dong European Conference on Computer Vision Workshops (ECCVW 2020) Over 390 citations Paper \| Codes