Jiuniu Wang

Joint PhD candidate

City University of Hong Kong

University of Chinese Academy of Sciences

Currently, I'm doing research on Distinctive Image Captioning with Prof. Antoni B. Chan at City University of Hong Kong.

I am a joint PhD student at the University of Chinese Academy of Sciences, advised by Prof. Yirong Wu. My current research interests are in computer vision, natural language processing, and deep neural networks.

My former background is in Electrical Engineering (B.Sc. hons. 2016 at Beijing Institute of Technology).

News

Selected Publications


Compare and Reweight: Distinctive Image Captioning Using Similar Images Sets

Proposing new metric for distinctiveness and new training strategy for distinctive image captioning.

J Wang, W Xu, Q Wang, A B. Chan, Compare and Reweight: Distinctive Image Captioning Using Similar Images Sets, ECCV (2020) Oral, [PDF, Project Page].


Attribute Prototype Network for Zero-Shot Learning

A ZSL network that can learn general attribute prototypes for various images.

W. Xu, Y. Xian, J. Wang, Z. Akata, B. Schiele, Attribute Prototype Network for Zero-Shot Learning, NeurIPS (2020), [PDF, Project Page].


Neighbours Matter: Image Captioning with Similar Images

Aggregating information over similar images is used to improve image captioning models.

Q Wang, J Wang, A B. Chan, S Huang, H Xiong, X Li, D Dou, Neighbours Matter: Image Captioning with Similar Images, BMVC (2020), [PDF].


Where is the Model Looking At --Concentrate and Explain the Network Attention

Investigating where is the model's attention when doing classification, and how can we guide the attention.

W. Xu, J Wang, Y Wang, G Xu, D Lin, Y Wu, Where is the Model Looking At --Concentrate and Explain the Network Attention, (IEEE Journal of Selected Topics in Signal Processing), [Link, PDF].


ASTRAL: Adversarial Trained LSTM-CNN for Named Entity Recognition

An adversarial trained method for Named Entity Recognition.

J Wang*, W. Xu*, X Fu, G Xu, Y Wu, ASTRAL: Adversarial Trained LSTM-CNN for Named Entity Recognition, (Knowledge-based Systems), (* = equal contribution), [Link, PDF].


SRQA: Synthetic Reader for Factoid Question Answering

A robust method for Question Answering.

J Wang*, W. Xu*, Y Wei, L Jin, Z Chen, G Xu, Y Wu, SRQA: Synthetic Reader for Factoid Question Answering, (Knowledge-based Systems), (* = equal contribution), [Link, PDF].


A3net: Adversarial-and-attention network for machine reading comprehension

Learning to recover high resolution remote sening images.

J Wang, X Fu, G Xu, Y Wu, Z Chen, Y Wei, L Jin, A3net: Adversarial-and-attention network for machine reading comprehension, NLPCC (2018) Oral ,[Link, PDF].


For a full list, have a look at my Google Scholar page.

Curriculum vitae