Experience

 
 
 
 
 

Research Intern

Pattern Recognition Center, WeChat AI, Tencent

Nov 2018 – Present Beijing, China

Natural Language Processing, Dialogue system, Knowledge Engineering, Deep Learning

Knowledge Powered Dialogue System :

  • Document Grounded Conversations. A long paper accepted by ACL 2019. A long paper accepted by EMNLP 2019 MRQA workshop.
  • Audio Visual Scene-aware Dialog. Rank 1st in DSTC8-AVSD Challenge. A paper accepted by AAAI2020 DSTC8 workshop.

Dialogue Coherence and Diversity : A long paper submitted to EMNLP 2020.

Dialogue State Tracking in Task-Oriented Dialogue: A long paper accepted by ACL2020.

 
 
 
 
 

AI Group Leader

Dian Group, Huazhong University of Science and Technology

Jul 2018 – Jun 2019 Wuhan, China
Founder of AI Group in Dian Group. AI Group contains 20+ undergraduate students who are interested in AI (natural language processing, computer vision, machine learning, deep learning, reinforcement learning). We cooperate with famous companies on commercial or research projects. We have published papers in ACL, CVPR.

  • AIbot: An Intelligent Customer Service Robot (Beibei Group).
  • AI Mooc: Build an intelligent question answering system for Mooc. We apply for the National Natural Science Funds.
  • AI in 5G: Using Deep Reinforcement Learning to optimize Network Access Control in 5G. (Intel Inc)
 
 
 
 
 

NLP Engineer Intern

Beibei Group

Mar 2018 – Nov 2018 Hangzhou, China
Team Leader of Intelligent Customer Service Robot. Beibei Group is the biggest e-commerce company for mother and baby in China. Lead a team of 12 undergraduate students to develop a stable, learnable and reliable intelligent customer service robot to replace the original commercial system. The resp onse accuracy and customer’s satisfaction of our system outperforms the original commercial one (Yib ot). Now AIbot serves hundreds of millions customers.
 
 
 
 
 

Research Intern

Wuhan National Laboratory for Optoelectronics, HUST

Oct 2017 – Jun 2018 Wuhan, China
Use deep reinforcement learning to tune cloud database automatically. Our system significantly outperforms the state-of-the-art tuning tools and DBA experts. And our paper is accepted by SIGMOD 2019.

Projects

*

Intelligent customer service system of Beibei Group

Automatic Cloud Database Tuning System using Deep Reinforcement Learning

Recent Publications

(2020). A Contextual Hierarchical Attention Network with Adaptive Objective for Dialogue State Tracking. ACL 2020 (Long Paper).

(2019). Bridging Text and Video: A Universal Multimodal Transformer for Video-Audio Scene-Aware Dialog (Rank 1st in DSTC8-AVSD Challenge). Proceedings of the 8th Dialog System Technology Challenge Workshop in AAAI2020.

(2019). Towards Multimodal Response Generation with Exemplar Augmentation and Curriculum Optimization. Arxiv.

(2019). Incremental Transformer with Deliberation Decoder for Document Grounded Conversations. ACL 2019 (Long Papers).

(2018). An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning. SIGMOD 2019.

Contact