Yizhe Zhang 张轶哲
 |
ContactSenior Researcher at Microsoft Research Company: yizhe.zhang (at) microsoft.com Personal: jeremy071242044 (at) gmail.com Room 3739, Building 99 One Microsoft way, Redmond, WA 98052, United States. |
Research interest
I am working on the interplays among natural language processing (NLP), deep generative models (DGM) and Reinforcement learning (RL). I have particular interests in text generation and conversational system. I also have broad interests for other natural language generation (NLG) tasks such as text style transfer, paraphrasing, question generation, summarization and machine translation. I am a Senior Program Committee (SPC) member for AAAI.
Specifically,
1). Large-scale transformer-based pretraining for text.
2). Non-autoregressive text generation.
3). Adversarial attacking for NLP.
4). Constraint/controllable text generation.
5). Reinforcement learning and self-playing for open-domain conversational agent.
6). Other interplays between deep generative models, RL, MCMC and NLP.
I will attend NeurIPS 2019 at Vancouver, please ping me for a chat if you are interested in above topics (esp. Dialogue, summarization and text generation evaluation) and looking for an internship.
Experience
Senior Researcher @ MSR NLP group. 2018 - present
Microsoft Research, Redmond, WA
Education
Ph.D on Machine Learning. 2013-2018
Advisor: Lawrence Carin
M.S. on Statistical Science. 2016-2018
Advisors: David Dunson, Scott Schmidler and Katherine Heller
Duke University, Durham, NC
Preprint
- Consistent Dialogue Generation with Self-supervised Feature Learning
Yizhe Zhang, Xiang Gao, Sungjin Lee, Chris Brockett, Michel Galley, Jianfeng Gao, Bill Dolan
- DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation
Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan
- Unsupervised Common Question Generation from Multiple Documents using Reinforced Contrastive Coordinator
Woon Sang Cho, Yizhe Zhang, Sudha Rao, Asli Celikyilmaz, Chenyan Xiong, Jianfeng Gao, Mengdi Wang, Bill Dolan
- INSET: Sentence Infilling with Inter-sentential Generative Pre-training
Yichen Huang, Yizhe Zhang, Oussama Elachqar, Yu Cheng
Selected conference and workshop publications
2019
- Complementary Auxiliary Classifiers for Label-Conditional Text Generation
Yuan Li, Chunyuan Li, Yizhe Zhang, Xiujun Li, Guoqing Zheng, Lawrence Carin, Jianfeng Gao — AAAI 2020
- Sequence Generation with Optimal-Transport-Enhanced Reinforcement Learning
Liqun Chen, Ke Bai, Chenyang Tao, Yizhe Zhang, Guoyin Wang, Wenlin Wang, Ricardo Henao, Lawrence Carin — AAAI 2020
- Structuring latent spaces for stylized response generation
Xiang Gao, Yizhe Zhang, Sungjin Lee, Michel Galley, Chris Brockett, Jianfeng Gao and Bill Dolan — EMNLP 2019
- Domain Adaptive Text Style Transfer
Dianqi Li, Yizhe Zhang, Zhe Gan, Yu Cheng, Chris Brockett, Ming-Ting Sun and Bill Dolan — EMNLP 2019
- Generating a Common Question from Multiple Documents using Multi-source Encoder-Decoder Models
Woon Sang Cho, Yizhe Zhang, Sudha Rao, Chris Brockett and Sungjin Lee — WNGT, EMNLP 2019
- Towards coherent and cohesive long-form text generation
Woon Sang Cho, Pengchuan Zhang, Yizhe Zhang, Xiujun Li, Michel Galley, Chris Brockett, Mengdi Wang, Jianfeng Gao — Workshop on Narrative Understanding, NAACL 2019
- Unsupervised Dialogue Spectrum Generation for Log Dialogue Ranking
Xinnuo Xu, Yizhe Zhang, Lars Liden and Sungjin Lee — SIGDIAL 2019 (Best paper nomination)
- Improving Textual Network Embedding with Global Attention via Optimal Transport
Liqun Chen, Guoyin Wang, Chenyang Tao, Dinghan Shen, Yizhe Zhang and Lawrence Carin — ACL 2019
- Towards Generating Long and Coherent Text with Multi-Level Latent Variable Models
Dinghan Shen, Asli Celikyilmaz, Yizhe Zhang, Liqun Chen, Xin Wang, Jianfeng Gao, Lawrence Carin — ACL 2019
- Jointly Optimizing Diversity and Relevance in Neural Response Generation.
Xiang Gao, Sungjin Lee, Yizhe Zhang, Chris Brockett, Michel Galley, Jianfeng Gao, Bill Dolan — NAACL 2019
- Improving Sequence-to-Sequence Learning via Optimal Transport.
Liqun Chen, Yizhe Zhang, Ruiyi Zhang, Chenyang Tao, Zhe Gan, Haichao Zhang, Bai Li, Dinghan Shen, Changyou Chen, Lawrence Carin — ICLR 2019
2018
- Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization.
Yizhe Zhang, Michel Galley, Jianfeng Gao, Zhe Gan, Xiujun Li, Chris Brockett, Bill Dolan — NIPS 2018
- Adversarial Text Generation via Feature-Mover's Distance.
Liqun Chen, Shuyang Dai, Chenyang Tao, Dinghan Shen, Zhe Gan, Haichao Zhang, Yizhe Zhang, Lawrence Carin — NIPS 2018
- Multi-Domain Joint Distribution Learning with Generative Adversarial Nets.
Yunchen Pu, Shuyang Dai, Zhe Gan, Weiyao Wang, Guoyin Wang, Yizhe Zhang, Ricardo Henao, Lawrence Carin — ICML 2018
- On Simple Word-Embedding-Based Models and Associated Pooling Mechanisms.
Dinghan Shen, Guoyin Wang, Wenlin Wang, Martin Renqiang Min, Qinliang Su, Yizhe Zhang, Chunyuan Li, Ricardo Henao and Lawrence Carin. — ACL 2018
- Joint Embedding of Words and Labels for Text Classification.
Guoyin Wang, Chunyuan Li, Wenlin Wang, Yizhe Zhang, Dinghan Shen, Xinyuan Zhang, Ricardo Henao and Lawrence Carin. — ACL 2018
- Deconvolutional Latent-Variable Model for Text Sequence Matching
Dinghan Shen, Yizhe Zhang, Ricardo Henao, Qinliang Su, Lawrence Carin. — AAAI 2018
- Zero-Shot Learning via Class-Conditioned Deep Generative Models
Wenlin Wang, Yunchen Pu, Vinay Kumar Verma, Kai Fan, Yizhe Zhang, Changyou Chen, Piyush Rai, Lawrence Carin. — AAAI 2018
2017
- Deconvolutional Paragraph Representation Learning [supplements] [code] [data]
Yizhe Zhang, Dinghan Shen, Guoyin Wang, Ricardo Henao, Zhe Gan, Lawrence Carin — NIPS 2017.
- Triangle Generative Adversarial Networks
Zhe Gan, Liqun Chen, Weiyao Wang, Yunchen Pu, Yizhe Zhang, Lawrence Carin — NIPS 2017.
- Stochastic Gradient Monomial Gamma Sampler [supplements] [code]
Yizhe Zhang, Changyou Chen, Zhe Gan, Ricardo Henao, Lawrence Carin. — ICML 2017.
- Adversarial Feature Matching for Text Generation [supplements] [code] [param] [data]
Yizhe Zhang, Zhe Gan, Kai Fan, Zhi Chen, Ricardo Henao, Lawrence Carin. — ICML 2017.
2016
- Towards Unifying Hamiltonian Monte Carlo and Slice Sampling [supplements] [code]
Yizhe Zhang, Xiangyu Wang, Changyou Chen, Lawrence Carin. — NIPS 2016.
- Distributed Bayesian Learning with Stochastic Gradient MCMC.
Changyou Chen, Nan Ding, Chunyuan Li, Yizhe Zhang, Lawrence Carin. — NIPS 2016.
- Generating Text via Adversarial Training.
Yizhe Zhang, Zhe Gan, Lawrence Carin. — Workshop on Adversarial Training, NIPS, 2016.
- Learning a Hybrid Architecture for Sequence Regression and Annotation. [supplements]
Yizhe Zhang, Ricardo Henao, Jianling Zhong, Lawrence Carin, Alexander Hartemink — AAAI 2016.
- Bayesian Dictionary Learning with Gaussian Processes and Sigmoid Belief Networks. [code]
Yizhe Zhang, Ricardo Henao, Chunyuan Li, Lawrence Carin. — IJCAI 2016.
- Triply Stochastic Variational Inference for Non-linear Beta Process Factor Analysis.
Kai Fan, Yizhe Zhang, Lawrence Carin, Katherine Heller. — ICDM 2016.
- Dynamic Poisson Factor Analysis [code]
Yizhe Zhang, Ricardo Henao, Lawrence Carin. — ICDM 2016
- Laplacian Hamiltonian Monte Carlo
Yizhe Zhang, Changyou Chen, Ricardo Henao, Lawrence Carin — ECML 2016.
2015
- Learning Dictionary with Spatial and Inter-dictionary Dependency.
Yizhe Zhang, Ricardo Henao, Chunyuan Li, Lawrence Carin. — Workshop on representation learning, NIPS, 2015.
Journal publications
- MOST+: a Motif Finding Approach Combining Genomic Sequence and Heterogeneous Genome-wide Signatures. [Source code in C++]
Yizhe Zhang, Yupeng He and Chaochun Wei. BMC Genomics, 2015.
- CRF-based Transcription Factor Binding Site Finding System. [Source code in C++]
Yupeng He, Yizhe Zhang, Guangyong Zheng and Chaochun Wei. BMC Genomics, 2012.
- Composition-based Classification of Short Metagenomic Sequences Elucidates the Landscapes of Taxonomic and Functional Enrichment of Microorganisms.
Jiemeng Liu, Haifeng Wang, Hongxing Yang, Yizhe Zhang, Jinfeng Wang, Fangqing Zhao and Ji Qi. Nucleic Acids Research, 2012.
Other projects
- Learning Infinite Mixture of Directed Acyclic Graphs
- Understanding Regulatory Element Topic via Relational Topic Modeling
Presentations
- Towards Improving the Efficiency and Scalability of MCMC inference
Ph.D. defense presentation at Duke. Unifying HMC and slice sampling and beyond.
Teaching
STA 561@Duke Probabilistic Machine Learning
STA 571@Duke Advanced Machine Learning
Professional services.
Meta-reviewer of AAAI (2019)
PC member of NIPS(2015-2019), ICML(2016-2019), AAAI(2016-2019), IJCAI(2016,2017), AISTATS(2018), ACL(2018-2019), EMNLP(2018-2019), NAACL(2019), SIAM(2018).
Students
- Guoyin Wang, Duke University, Research intern 2019
- Ziyu Yao, Ohio State University, Research intern 2018
- Woon Sang Cho, Princeton University, Research intern 2018 (co-mentoring)
- Dinghan Shen, Duke University, Research intern 2018 (co-mentoring)
News
[Aug. 2019] Serving as local web chairs for ACL 2020.
[Aug. 2019] Our SIGDIAL paper is nominated as Best paper candidates.
[May. 2019] Two papers accepted by ACL 2019
[May. 2019] Going to New Orlean for ICLR 2019
[Oct. 2018] Going to Montreal for NeurIPS 2018 this December. Reach out to me if you will also go!
[Aug. 2018] Our recent papers “Adversarial Text Generation via Feature-Mover’s Distance” and “Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization” are accepted by NIPS 2018.
[Mar. 2018] Joined Microsoft Research as a full-time researcher.