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 Kaitao Song

Kaitao Song

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions


Feb 24, 2021
Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao

* 13 pages, 6 figures, technical report 
  Access Paper or Ask Questions

SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint


Dec 09, 2020
Zhonghao Sheng, Kaitao Song, Xu Tan, Yi Ren, Wei Ye, Shikun Zhang, Tao Qin

Figure 1 for SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint
Figure 2 for SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint
Figure 3 for SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint
Figure 4 for SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint

  Access Paper or Ask Questions

Neural Machine Translation with Error Correction


Jul 21, 2020
Kaitao Song, Xu Tan, Jianfeng Lu

Figure 1 for Neural Machine Translation with Error Correction
Figure 2 for Neural Machine Translation with Error Correction
Figure 3 for Neural Machine Translation with Error Correction
Figure 4 for Neural Machine Translation with Error Correction
* Accepted by IJCAI 2020 
  Access Paper or Ask Questions

LightPAFF: A Two-Stage Distillation Framework for Pre-training and Fine-tuning


Apr 27, 2020
Kaitao Song, Hao Sun, Xu Tan, Tao Qin, Jianfeng Lu, Hongzhi Liu, Tie-Yan Liu

Figure 1 for LightPAFF: A Two-Stage Distillation Framework for Pre-training and Fine-tuning
Figure 2 for LightPAFF: A Two-Stage Distillation Framework for Pre-training and Fine-tuning
Figure 3 for LightPAFF: A Two-Stage Distillation Framework for Pre-training and Fine-tuning
Figure 4 for LightPAFF: A Two-Stage Distillation Framework for Pre-training and Fine-tuning

  Access Paper or Ask Questions

MPNet: Masked and Permuted Pre-training for Language Understanding


Apr 20, 2020
Kaitao Song, Xu Tan, Tao Qin, Jianfeng Lu, Tie-Yan Liu

Figure 1 for MPNet: Masked and Permuted Pre-training for Language Understanding
Figure 2 for MPNet: Masked and Permuted Pre-training for Language Understanding
Figure 3 for MPNet: Masked and Permuted Pre-training for Language Understanding
Figure 4 for MPNet: Masked and Permuted Pre-training for Language Understanding

  Access Paper or Ask Questions

MASS: Masked Sequence to Sequence Pre-training for Language Generation


May 13, 2019
Kaitao Song, Xu Tan, Tao Qin, Jianfeng Lu, Tie-Yan Liu

Figure 1 for MASS: Masked Sequence to Sequence Pre-training for Language Generation
Figure 2 for MASS: Masked Sequence to Sequence Pre-training for Language Generation
Figure 3 for MASS: Masked Sequence to Sequence Pre-training for Language Generation
Figure 4 for MASS: Masked Sequence to Sequence Pre-training for Language Generation
* Accepted by ICML 2019 
  Access Paper or Ask Questions

Generating Adversarial Examples With Conditional Generative Adversarial Net


Mar 18, 2019
Ping Yu, Kaitao Song, Jianfeng Lu

Figure 1 for Generating Adversarial Examples With Conditional Generative Adversarial Net
Figure 2 for Generating Adversarial Examples With Conditional Generative Adversarial Net
Figure 3 for Generating Adversarial Examples With Conditional Generative Adversarial Net
Figure 4 for Generating Adversarial Examples With Conditional Generative Adversarial Net
* 6 pages 
  Access Paper or Ask Questions

Hybrid Self-Attention Network for Machine Translation


Nov 01, 2018
Kaitao Song, Tan Xu, Furong Peng, Jianfeng Lu

Figure 1 for Hybrid Self-Attention Network for Machine Translation
Figure 2 for Hybrid Self-Attention Network for Machine Translation
Figure 3 for Hybrid Self-Attention Network for Machine Translation
Figure 4 for Hybrid Self-Attention Network for Machine Translation

  Access Paper or Ask Questions

Double Path Networks for Sequence to Sequence Learning


Jul 04, 2018
Kaitao Song, Xu Tan, Di He, Jianfeng Lu, Tao Qin, Tie-Yan Liu

Figure 1 for Double Path Networks for Sequence to Sequence Learning
Figure 2 for Double Path Networks for Sequence to Sequence Learning
Figure 3 for Double Path Networks for Sequence to Sequence Learning
Figure 4 for Double Path Networks for Sequence to Sequence Learning
* 11 pages, to appear in COLING 2018 
  Access Paper or Ask Questions