Bowen Tan (谭博文)

Ph.D. Student
Language Technologies Institute (LTI)
Carnegie Mellon University (CMU)

Email: btan2 (at) andrew.cmu.edu

Research Interests

Machine Learning (ML) and Natural Language Processing (NLP) in general, including text generation, knowledge graph, dialogue systems, reinforcement learning, graph neural networks, and composable ML systems. Specially, I am interested in the systemization of machine learning paradigms, to enable training AI agents with arbitrary combinations of experiences (e.g., data examples, structured knowledge, adversaries, auxiliary models).

Education

  • 2021, M.S., Language Technologies Institute, Carnegie Mellon University
    Advisor: Eric P. Xing and Zhiting Hu
  • 2019, B.S., ACM Honors Class at Zhiyuan College, Shanghai Jiao Tong University, China
    Advisor: Kai Yu

Publications

  • Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation
    Bowen Tan*, Mingkai Deng* (equal contrib.), Zhengzhong Liu, Eric P Xing, Zhiting Hu
    EMNLP 2021 [arXiv] [Code]
  • Text Generation with Efficient (Soft) Q-Learning
    Han Guo, Bowen Tan, Zhengzhong Liu, Eric P Xing, Zhiting Hu
    Pre-print 2021 [arXiv] [Code]
  • Progressive Generation of Long Text
    Bowen Tan, Zichao Yang, Maruan AI-Shedivat, Eric P. Xing, Zhiting Hu
    NAACL 2021 [arXiv] [Slides] [Code]
  • Summarizing Text on Any Aspects: A Knowledge-Informed Weakly-Supervised Approach
    Bowen Tan, Lianhui Qin, Eric Xing and Zhiting Hu
    EMNLP 2020 [arXiv] [Slides] [Code]
  • Learning Data Manipulation for Augmentation and Weighting
    Bowen Tan*, Zhiting Hu* (equal contrib.), Ruslan Salakhutdinov, Tom Mitchell, Eric P. Xing
    NeurIPS 2019 [arXiv] [Code]
  • Connecting the Dots between MLE and RL for Sequence Generation
    Bowen Tan*, Zhiting Hu* (equal contrib.), Zichao Yang, Ruslan Salakhutdinov, Eric Xing
    Pre-print 2019 [arXiv] [Code] [Blog] [Blog (in Chinese)]
    (Best paper on ICLR2019 drlStructPred workshop)
  • Structured Dialogue Policy with Graph Neural Networks
    Bowen Tan*, Lu Chen* (equal contrib.), Sishan Long, Kai Yu
    COLING 2018 [arXiv]
    (selected as "Area Chair Favorites".)
  • Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
    Zhiting Hu, Haoran Shi, Bowen Tan, Wentao Wang, Zichao Yang, Tiancheng Zhao, Junxian He, Lianhui Qin, Di Wang, Xuezhe Ma, Zhengzhong Liu, Xiaodan Liang, Wangrong Zhu, Devendra Singh Sachan, Eric P. Xing
    ACL 2019 (demo paper) [arXiv] [slides] [Website] [GitHub (TensorFlow)] [GitHub (PyTorch)] [Blog] [Blog (in Chinese)]
    (Best demo paper nomination)
  • On the Generation of Medical Dialogs for COVID-19
    Meng Zhou, Zechen Li, Bowen Tan, Guangtao Zeng, Wenmian Yang, Xuehai He, Zeqian Ju, Subrato Chakravorty, Shu Chen, Xingyi Yang, Yichen Zhang, Qingyang Wu, Zhou Yu, Eric Xing, Pengtao Xie
    ACL 2021
  • AgentGraph: Towards Universal Dialogue Management with Structured Deep Reinforcement Learning
    Lu Chen, Zhi Chen, Bowen Tan, Sishan Long, Milica Gasic, Kai Yu
    IEEE/ACM Transactions on Audio, Speech, and Language Processing 2019 [arXiv]
  • Schema-Guided Multi-Domain Dialogue State Tracking with Graph Attention Neural Networks
    Lu Chen, Boer Lv, Chi Wang, Su Zhu, Bowen Tan, Kai Yu
    AAAI 2020 [arXiv]
  • Policy Adaptation for Deep Reinforcement Learning-based Dialogue Management
    Lu Chen, Cheng Chang, Zhi Chen, Bowen Tan, Milica Gasic, Kai Yu
    ICASSP 2018 [arXiv]

Awards & Honors (Selected)

  • ACL 2019 Best Demo Paper Nomination, 2019.
  • ICLR 2019 drlStructPred Workshop Best Paper, 2019.
  • COLING 2018 "Area Chair Favorites", 2018.
  • RongChang Scholarship, Shanghai Jiao Tong University, 2017 & 2018.
  • Zhiyuan Outstanding Scholarship, Shanghai Jiao Tong University, 2019.
  • Zhiyuan Honor Degree of Bachelor, Shanghai Jiao Tong University, 2019.
  • Outstanding Graguate, Shanghai Jiao Tong University, 2019.
  • Second Runner-up, ACM-ICPC Reginal Nha-Trang, 2016.
  • Gold Medal, ACM-ICPC China-Final, 2016.
  • Gold Medal, ACM-ICPC Reginal Beijing, 2015.