Short-bio
Phd candidate at Harbin Institute of Technology, HIT. Advised by Prof. Haifeng Wang and Prof. Wanxiang Che.
From November 2014 to June 2015 I was a research assistant at NUS and received my bachelor of computer science degree from Harbin Institute of Technology.
Research Interests
- Knowledge-driven Proactive Open-Domain Dialogue. I am pondering how to endow chatbot with proactivity. Maybe a brain (graph based) is needed, so I am trying to build it, and we will see what will happen...
Selected Publications and Notes
- Jun Xu, Zeyang Lei, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che: Discovering Dialog Structure Graph for Coherent Dialog Generation. ACL2021-Main Conference PDF
- Jun Xu, Zeyang Lei, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che, Jizhou Huang, Ting Liu: Coherent Dialog Generation with Query Graph. TALLIP
- Jun Xu, Zeyang Lei, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che: Enhancing Dialog Coherence with Event Graph Grounded Content Planning. IJCAI2020 PDF
- Jun Xu, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che: Conversational Graph Grounded Policy Learning for Open-Domain Conversation Generation. ACL2020-Main Conference PDF CG
- Jun Xu, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che: Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation. AAAI2020 PDF
- Jun Xu, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che: A Key-Phrase Aware End2end Neural Response Generation Model. NLPCC2019 PDF PPT
- Chaotao Chen, Jinhua Peng, Fan Wang, Jun Xu, Hua Wu: Generating Multiple Diverse Responses with Multi-Mapping and Posterior Mapping Selection. IJCAI2019 PDF
- Jiang Guo, Wanxiang Che, Haifeng Wang, Ting Liu, Jun Xu: A Unified Architecture for Semantic Role Labeling and Relation Classification. COLING2016
- Tang, Yanbin, Junbin Fang, K. P. Chow, S. M. Yiu, Jun Xu, Bo Feng, Qiong Li, and Qi Han. "Recovery of heavily fragmented JPEG files." Digital Investigation 18 (2016): S108-S117. PDF
- D. Jiang, S. Wu, G. Chen, B.C. Ooi, K.-L. Tan, J. Xu: epiC: an extensible and scalable system for processing Big Data. VLDB Journal- Special issue of VLDB'14 best papers, 2015.
- Word2Vec Explained PPT
- Sequence to Sequence Learning: A Feature's Perspective PPT
Research and Intern Experience
- 2014.10 - 2015.06: Research Assistant at School of Computing, NUS
Supervised by Prof. Beng Chin OOI and Dr. David Jiang. Mainly focuse on distributed database recovering and business intelligence cube.- 2014.07 - 2014.09: Summer Intern at Department of Natural Language Processing, Baidu
Supervised by Xinyu Zhang. Working on word embedding.- 2013.09 - 2014.06: Research Assistant at Department of Computer Science, HIT
Supervised by Prof. Hongzhi Wang. Focused on Big data and completed a paper calculates bi-simulation of graph(with billions of edges) based on the GPS system.- 2013.07 - 2013.09: Research Assistant at Department of Computer Science, HKU
Supervised by Dr. S.M.Yiu. Focused on File Carving and completed one paper. - 2014.07 - 2014.09: Summer Intern at Department of Natural Language Processing, Baidu
Projects and Achievements
- Business Intelligence Cube(Initial developer, Query engine, RESTFul, Compressor)
- Distributed database recovering(Zookeeper, LevelDB, HDFS)
- Incremental word2vec
- Silver Awards in The ACM Programming Contest of HIT
- Won the First Prize (also as The First editor) in China Undergraduate Mathematical Contest in Modeling
- Excellent Graduate of Harbin Institute of Technology
- National Scholarship
- Excellent Cadre of Students and Jianke Elite Scholarship
- Excellent Achievements Scholarship and Xerox Corporation Scholarship
Technical Skills
- Programming Languages: C/C++, Python, Shell, Java, LaTeX
- Experiences: Hadoop, Spark, Linux, Git, Vim