![]() |
Tianxiao Shen 沈添笑Paul G. Allen School of Computer Science & EngineeringUniversity of Washington 185 E Stevens Way NE, Seattle, WA 98195 Office: CSE 328 Email: stx [at] cs [dot] washington [dot] edu [Google Scholar] [Github] |
I am a postdoctoral scholar at the University of Washington, working with Yejin Choi and Zaid Harchaoui. Previously, I received my PhD from MIT, advised by Regina Barzilay and Tommi Jaakkola. Before that, I did my undergrad at Tsinghua University, where I was a member of the Yao Class.
I have broad interests in natural language processing and machine learning, with a focus on text generation. My research goal is to advance text generation models and algorithms that can be employed in real-world contexts to facilitate, curate, and inspire human writing.
(*: Equal contribution) |
Generating Sequences by Learning to [Self-]Correct |
Sean Welleck*, Ximing Lu*, Peter West, Faeze Brahman, Tianxiao Shen, Daniel Khashabi, Yejin Choi |
ICLR 2023 |
[paper] |
Controlling Neural Language Generation |
Tianxiao Shen |
PhD thesis, MIT 2022 |
[pdf] |
Text Style Transfer with Confounders |
Tianxiao Shen, Regina Barzilay, Tommi Jaakkola |
Preprint 2022 |
[paper] [slides] |
Controlling Directions Orthogonal to a Classifier |
Yilun Xu, Hao He, Tianxiao Shen, Tommi Jaakkola |
ICLR 2022 Spotlight |
[paper] |
Blank Language Models |
Tianxiao Shen*, Victor Quach*, Regina Barzilay, Tommi Jaakkola |
EMNLP 2020 |
[paper] [slides] [code] [video] |
Educating Text Autoencoders: Latent Representation Guidance via Denoising |
Tianxiao Shen, Jonas Mueller, Regina Barzilay, Tommi Jaakkola |
ICML 2020 |
[paper] [slides] [code] [video] |
Learning to Make Generalizable and Diverse Predictions for Retrosynthesis |
Benson Chen, Tianxiao Shen, Tommi Jaakkola, Regina Barzilay |
Preprint 2019 |
[paper] |
Mixture Models for Diverse Machine Translation: Tricks of the Trade |
Tianxiao Shen*, Myle Ott*, Michael Auli, Marc'Aurelio Ranzato |
ICML 2019 Long talk |
[paper] [slides] [code] [video] |
Style Transfer from Non-Parallel Text by Cross-Alignment |
Tianxiao Shen, Tao Lei, Regina Barzilay, Tommi Jaakkola |
NIPS 2017 Spotlight |
[paper] [slides] [code] [video] |
Making Dependency Labeling Simple, Fast and Accurate |
Tianxiao Shen, Tao Lei, Regina Barzilay |
NAACL 2016 |
[paper] [code] |
Oct 2021 | NetMIT. Text Style Transfer with Confounders |
Sep 2021 | IBM. Text Style Transfer with Confounders |
Dec 2020 | AI TIME. Blank Language Models |
Aug 2020 | Princeton NLP. Blank Language Models |
Jul 2020 | MIT Computational Psycholinguistics Lab. Blank Language Models |
Jul 2020 | MIT MLTea. Educating Text Autoencoders: Latent Representation Guidance via Denoising |
Sep 2019 | AI Research Week, MIT-IBM Watson AI Lab. Mixture Models for Diverse Machine Translation |
Jul 2019 | IIIS-Haihua Frontier Seminar Series, Tsinghua University. Latent Variable Language Modeling |
Mar 2018 | Harvard Applied Statistics Seminar. Language Style Transfer |
Mar 2018 | NetMIT. Language Style Transfer |
Dec 2017 | Twitter Cambridge. Language Style Transfer |
Nov 2017 | Harvard NLP. Language Style Transfer |
Nov 2017 | Guest lecture at MIT 6.864 Advanced Natural Language Processing. Language Style Transfer |
Oct 2017 | MIT Machine Learning Tea. Language Style Transfer |
6/2020 - 9/2020 | Visiting student at Princeton NLP Group |
6/2018 - 8/2018 | Research intern at Facebook AI Research |
2/2016 - 7/2016 | Research intern at Microsoft Research Asia |
6/2014 - 12/2014 | Research intern at Microsoft Research Asia |
Fall 2022 | Organizer for Diffusion Models Reading Group at UW |
Spring 2020 | Teaching assistant for 6.883 Modeling with Machine Learning at MIT |
I am widely interested in philosophy. Here are my writing samples: