Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Junxian He

Junxian He

Towards a Unified View of Parameter-Efficient Transfer Learning


Oct 08, 2021
Junxian He, Chunting Zhou, Xuezhe Ma, Taylor Berg-Kirkpatrick, Graham Neubig

Figure 1 for Towards a Unified View of Parameter-Efficient Transfer Learning
Figure 2 for Towards a Unified View of Parameter-Efficient Transfer Learning
Figure 3 for Towards a Unified View of Parameter-Efficient Transfer Learning
Figure 4 for Towards a Unified View of Parameter-Efficient Transfer Learning
* under review 
  Access Paper or Ask Questions

Capturing Structural Locality in Non-parametric Language Models


Oct 06, 2021
Frank F. Xu, Junxian He, Graham Neubig, Vincent J. Hellendoorn

Figure 1 for Capturing Structural Locality in Non-parametric Language Models
Figure 2 for Capturing Structural Locality in Non-parametric Language Models
Figure 3 for Capturing Structural Locality in Non-parametric Language Models
Figure 4 for Capturing Structural Locality in Non-parametric Language Models

  Access Paper or Ask Questions

Dependency Induction Through the Lens of Visual Perception


Sep 20, 2021
Ruisi Su, Shruti Rijhwani, Hao Zhu, Junxian He, Xinyu Wang, Yonatan Bisk, Graham Neubig

Figure 1 for Dependency Induction Through the Lens of Visual Perception
Figure 2 for Dependency Induction Through the Lens of Visual Perception
Figure 3 for Dependency Induction Through the Lens of Visual Perception
Figure 4 for Dependency Induction Through the Lens of Visual Perception
* Accepted to CoNLL 2021 
  Access Paper or Ask Questions

Efficient Nearest Neighbor Language Models


Sep 09, 2021
Junxian He, Graham Neubig, Taylor Berg-Kirkpatrick

Figure 1 for Efficient Nearest Neighbor Language Models
Figure 2 for Efficient Nearest Neighbor Language Models
Figure 3 for Efficient Nearest Neighbor Language Models
Figure 4 for Efficient Nearest Neighbor Language Models
* EMNLP 2021 
  Access Paper or Ask Questions

CTRLsum: Towards Generic Controllable Text Summarization


Dec 08, 2020
Junxian He, Wojciech Kryściński, Bryan McCann, Nazneen Rajani, Caiming Xiong

Figure 1 for CTRLsum: Towards Generic Controllable Text Summarization
Figure 2 for CTRLsum: Towards Generic Controllable Text Summarization
Figure 3 for CTRLsum: Towards Generic Controllable Text Summarization
Figure 4 for CTRLsum: Towards Generic Controllable Text Summarization
* Preprint 
  Access Paper or Ask Questions

On the Sentence Embeddings from Pre-trained Language Models


Nov 02, 2020
Bohan Li, Hao Zhou, Junxian He, Mingxuan Wang, Yiming Yang, Lei Li

Figure 1 for On the Sentence Embeddings from Pre-trained Language Models
Figure 2 for On the Sentence Embeddings from Pre-trained Language Models
Figure 3 for On the Sentence Embeddings from Pre-trained Language Models
Figure 4 for On the Sentence Embeddings from Pre-trained Language Models
* EMNLP 2020 
  Access Paper or Ask Questions

Learning Sparse Prototypes for Text Generation


Jun 29, 2020
Junxian He, Taylor Berg-Kirkpatrick, Graham Neubig

Figure 1 for Learning Sparse Prototypes for Text Generation
Figure 2 for Learning Sparse Prototypes for Text Generation
Figure 3 for Learning Sparse Prototypes for Text Generation
Figure 4 for Learning Sparse Prototypes for Text Generation
* 8 pages of main content 
  Access Paper or Ask Questions

A Probabilistic Formulation of Unsupervised Text Style Transfer


Mar 11, 2020
Junxian He, Xinyi Wang, Graham Neubig, Taylor Berg-Kirkpatrick

Figure 1 for A Probabilistic Formulation of Unsupervised Text Style Transfer
Figure 2 for A Probabilistic Formulation of Unsupervised Text Style Transfer
Figure 3 for A Probabilistic Formulation of Unsupervised Text Style Transfer
Figure 4 for A Probabilistic Formulation of Unsupervised Text Style Transfer
* ICLR 2020 conference paper (spotlight). The first two authors contributed equally 
  Access Paper or Ask Questions

Revisiting Self-Training for Neural Sequence Generation


Sep 30, 2019
Junxian He, Jiatao Gu, Jiajun Shen, Marc'Aurelio Ranzato

Figure 1 for Revisiting Self-Training for Neural Sequence Generation
Figure 2 for Revisiting Self-Training for Neural Sequence Generation
Figure 3 for Revisiting Self-Training for Neural Sequence Generation
Figure 4 for Revisiting Self-Training for Neural Sequence Generation

  Access Paper or Ask Questions

The Source-Target Domain Mismatch Problem in Machine Translation


Sep 28, 2019
Jiajun Shen, Peng-Jen Chen, Matt Le, Junxian He, Jiatao Gu, Myle Ott, Michael Auli, Marc'Aurelio Ranzato

Figure 1 for The Source-Target Domain Mismatch Problem in Machine Translation
Figure 2 for The Source-Target Domain Mismatch Problem in Machine Translation
Figure 3 for The Source-Target Domain Mismatch Problem in Machine Translation
Figure 4 for The Source-Target Domain Mismatch Problem in Machine Translation

  Access Paper or Ask Questions

A Surprisingly Effective Fix for Deep Latent Variable Modeling of Text


Sep 02, 2019
Bohan Li, Junxian He, Graham Neubig, Taylor Berg-Kirkpatrick, Yiming Yang

Figure 1 for A Surprisingly Effective Fix for Deep Latent Variable Modeling of Text
Figure 2 for A Surprisingly Effective Fix for Deep Latent Variable Modeling of Text
Figure 3 for A Surprisingly Effective Fix for Deep Latent Variable Modeling of Text
Figure 4 for A Surprisingly Effective Fix for Deep Latent Variable Modeling of Text
* EMNLP 2019 short paper. The first two authors contributed equally 
  Access Paper or Ask Questions

Cross-Lingual Syntactic Transfer through Unsupervised Adaptation of Invertible Projections


Jun 17, 2019
Junxian He, Zhisong Zhang, Taylor Berg-Kiripatrick, Graham Neubig

Figure 1 for Cross-Lingual Syntactic Transfer through Unsupervised Adaptation of Invertible Projections
Figure 2 for Cross-Lingual Syntactic Transfer through Unsupervised Adaptation of Invertible Projections
Figure 3 for Cross-Lingual Syntactic Transfer through Unsupervised Adaptation of Invertible Projections
Figure 4 for Cross-Lingual Syntactic Transfer through Unsupervised Adaptation of Invertible Projections
* ACL 2019 long paper 
  Access Paper or Ask Questions

Choosing Transfer Languages for Cross-Lingual Learning


Jun 07, 2019
Yu-Hsiang Lin, Chian-Yu Chen, Jean Lee, Zirui Li, Yuyan Zhang, Mengzhou Xia, Shruti Rijhwani, Junxian He, Zhisong Zhang, Xuezhe Ma, Antonios Anastasopoulos, Patrick Littell, Graham Neubig

Figure 1 for Choosing Transfer Languages for Cross-Lingual Learning
Figure 2 for Choosing Transfer Languages for Cross-Lingual Learning
Figure 3 for Choosing Transfer Languages for Cross-Lingual Learning
Figure 4 for Choosing Transfer Languages for Cross-Lingual Learning
* Proceedings of ACL 2019 
  Access Paper or Ask Questions

Lagging Inference Networks and Posterior Collapse in Variational Autoencoders


Jan 28, 2019
Junxian He, Daniel Spokoyny, Graham Neubig, Taylor Berg-Kirkpatrick

Figure 1 for Lagging Inference Networks and Posterior Collapse in Variational Autoencoders
Figure 2 for Lagging Inference Networks and Posterior Collapse in Variational Autoencoders
Figure 3 for Lagging Inference Networks and Posterior Collapse in Variational Autoencoders
Figure 4 for Lagging Inference Networks and Posterior Collapse in Variational Autoencoders
* ICLR 2019 Conference Paper 
  Access Paper or Ask Questions

Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation


Sep 04, 2018
Zhiting Hu, Haoran Shi, Zichao Yang, Bowen Tan, Tiancheng Zhao, Junxian He, Wentao Wang, Xingjiang Yu, Lianhui Qin, Di Wang, Xuezhe Ma, Hector Liu, Xiaodan Liang, Wanrong Zhu, Devendra Singh Sachan, Eric P. Xing

Figure 1 for Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
Figure 2 for Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
Figure 3 for Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
Figure 4 for Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation
* 14 pages; Github: https://github.com/asyml/texar 
  Access Paper or Ask Questions

Unsupervised Learning of Syntactic Structure with Invertible Neural Projections


Aug 28, 2018
Junxian He, Graham Neubig, Taylor Berg-Kirkpatrick

Figure 1 for Unsupervised Learning of Syntactic Structure with Invertible Neural Projections
Figure 2 for Unsupervised Learning of Syntactic Structure with Invertible Neural Projections
Figure 3 for Unsupervised Learning of Syntactic Structure with Invertible Neural Projections
Figure 4 for Unsupervised Learning of Syntactic Structure with Invertible Neural Projections
* EMNLP 2018 
  Access Paper or Ask Questions

StructVAE: Tree-structured Latent Variable Models for Semi-supervised Semantic Parsing


Jun 20, 2018
Pengcheng Yin, Chunting Zhou, Junxian He, Graham Neubig

Figure 1 for StructVAE: Tree-structured Latent Variable Models for Semi-supervised Semantic Parsing
Figure 2 for StructVAE: Tree-structured Latent Variable Models for Semi-supervised Semantic Parsing
Figure 3 for StructVAE: Tree-structured Latent Variable Models for Semi-supervised Semantic Parsing
Figure 4 for StructVAE: Tree-structured Latent Variable Models for Semi-supervised Semantic Parsing
* ACL 2018 
  Access Paper or Ask Questions

Efficient Correlated Topic Modeling with Topic Embedding


Jul 01, 2017
Junxian He, Zhiting Hu, Taylor Berg-Kirkpatrick, Ying Huang, Eric P. Xing

Figure 1 for Efficient Correlated Topic Modeling with Topic Embedding
Figure 2 for Efficient Correlated Topic Modeling with Topic Embedding
Figure 3 for Efficient Correlated Topic Modeling with Topic Embedding
Figure 4 for Efficient Correlated Topic Modeling with Topic Embedding
* KDD 2017 oral. The first two authors contributed equally 
  Access Paper or Ask Questions

Text Network Exploration via Heterogeneous Web of Topics


Oct 02, 2016
Junxian He, Ying Huang, Changfeng Liu, Jiaming Shen, Yuting Jia, Xinbing Wang

Figure 1 for Text Network Exploration via Heterogeneous Web of Topics
Figure 2 for Text Network Exploration via Heterogeneous Web of Topics
Figure 3 for Text Network Exploration via Heterogeneous Web of Topics
Figure 4 for Text Network Exploration via Heterogeneous Web of Topics
* 8 pages 
  Access Paper or Ask Questions