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 Zihang Dai

CoAtNet: Marrying Convolution and Attention for All Data Sizes

Jun 09, 2021
Zihang Dai, Hanxiao Liu, Quoc V. Le, Mingxing Tan

  Access Paper or Ask Questions

Pay Attention to MLPs

Jun 01, 2021
Hanxiao Liu, Zihang Dai, David R. So, Quoc V. Le

  Access Paper or Ask Questions

Unsupervised Parallel Corpus Mining on Web Data

Sep 18, 2020
Guokun Lai, Zihang Dai, Yiming Yang

  Access Paper or Ask Questions

Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing

Jun 05, 2020
Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le

  Access Paper or Ask Questions

Meta Pseudo Labels

Apr 23, 2020
Hieu Pham, Qizhe Xie, Zihang Dai, Quoc V. Le

* Preprint 

  Access Paper or Ask Questions

A Mutual Information Maximization Perspective of Language Representation Learning

Nov 26, 2019
Lingpeng Kong, Cyprien de Masson d'Autume, Wang Ling, Lei Yu, Zihang Dai, Dani Yogatama

* 12 pages, 3 figures 

  Access Paper or Ask Questions

XLNet: Generalized Autoregressive Pretraining for Language Understanding

Jun 19, 2019
Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le

* Pretrained models and code are available at 

  Access Paper or Ask Questions

Unsupervised Data Augmentation

Apr 29, 2019
Qizhe Xie, Zihang Dai, Eduard Hovy, Minh-Thang Luong, Quoc V. Le

  Access Paper or Ask Questions

Re-examination of the Role of Latent Variables in Sequence Modeling

Feb 04, 2019
Zihang Dai, Guokun Lai, Yiming Yang, Shinjae Yoo

* Code available at 

  Access Paper or Ask Questions

Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context

Jan 18, 2019
Zihang Dai, Zhilin Yang, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov

* Code and pretrained models are available at 

  Access Paper or Ask Questions

Characterizing and Avoiding Negative Transfer

Nov 24, 2018
Zirui Wang, Zihang Dai, Barnabás Póczos, Jaime Carbonell

  Access Paper or Ask Questions

Fast and Simple Mixture of Softmaxes with BPE and Hybrid-LightRNN for Language Generation

Sep 25, 2018
Xiang Kong, Qizhe Xie, Zihang Dai, Eduard Hovy

  Access Paper or Ask Questions

SwitchOut: an Efficient Data Augmentation Algorithm for Neural Machine Translation

Aug 28, 2018
Xinyi Wang, Hieu Pham, Zihang Dai, Graham Neubig

* Accepted as a short paper at the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018) 

  Access Paper or Ask Questions

Large-scale Cloze Test Dataset Created by Teachers

Aug 28, 2018
Qizhe Xie, Guokun Lai, Zihang Dai, Eduard Hovy

* EMNLP 2018 

  Access Paper or Ask Questions

From Credit Assignment to Entropy Regularization: Two New Algorithms for Neural Sequence Prediction

Apr 29, 2018
Zihang Dai, Qizhe Xie, Eduard Hovy

* ACL 2018 

  Access Paper or Ask Questions

Breaking the Softmax Bottleneck: A High-Rank RNN Language Model

Mar 02, 2018
Zhilin Yang, Zihang Dai, Ruslan Salakhutdinov, William W. Cohen

* ICLR Oral 2018 

  Access Paper or Ask Questions

Controllable Invariance through Adversarial Feature Learning

Jan 29, 2018
Qizhe Xie, Zihang Dai, Yulun Du, Eduard Hovy, Graham Neubig

* NIPS 2017 

  Access Paper or Ask Questions

Good Semi-supervised Learning that Requires a Bad GAN

Nov 03, 2017
Zihang Dai, Zhilin Yang, Fan Yang, William W. Cohen, Ruslan Salakhutdinov

* NIPS 2017 camera ready 

  Access Paper or Ask Questions

An Interpretable Knowledge Transfer Model for Knowledge Base Completion

May 03, 2017
Qizhe Xie, Xuezhe Ma, Zihang Dai, Eduard Hovy

* Accepted by ACL 2017. Minor update 

  Access Paper or Ask Questions

Calibrating Energy-based Generative Adversarial Networks

Feb 24, 2017
Zihang Dai, Amjad Almahairi, Philip Bachman, Eduard Hovy, Aaron Courville

* ICLR 2017 camera ready 

  Access Paper or Ask Questions

CFO: Conditional Focused Neural Question Answering with Large-scale Knowledge Bases

Jul 04, 2016
Zihang Dai, Lei Li, Wei Xu

* Accepted by ACL 2016 

  Access Paper or Ask Questions