awesome-text-generation

A curated list (as of EMNLP 2018, will update to NeurIPS18 soon) of recent awesome text generation model and their application. Inspired by awesome-architecture-search, awesome-adversarial-machine-learining and awesome-deep-learning-paper. Please help to contribute if you find some important works are missing.

• Model
  • GAN based
  • VAE based
  • Autoencoder based
  • Reinforcement learning based
  • Alternative decode objective
  • Tool and others
• Applications
  • Reinforcement Learning based text generation
    • Image to text
    • Stylistic Text
    • (Visual) Dialogue
    • Other
  • VAE based
    • (Visual) Dialogue
    • Stylistic Text
  • GAN based (Adversarial Learining)
    • Image to text
    • Stylistic Text
    • Other
  • Other

Model

GAN based

• GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution [pdf]
  • Matt J. Kusner, José Miguel Hernández-Lobato ICLR 2018
• Adversarial Feature Matching for Text Generation [pdf] [code]
  • Yizhe Zhang, Zhe Gan, Kai Fan, Zhi Chen, Ricardo Henao, Dinghan Shen, Lawrence Carin ICML 2017
• Improved Training of Wasserstein GANs [pdf][code]
  • Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, Aaron Courville NIPS 2017
• Sobolev GAN [pdf][code]
  • Youssef Mroueh, Chun-Liang Li, Tom Sercu, Anant Raj, Yu Cheng ICLR 2018

VAE based

• Spherical Latent Spaces for Stable Variational Autoencoders. [pdf][code]
  • Jiacheng Xu and Greg Durrett EMNLP 2018
• Semi-Amortized Variational Autoencoders [pdf][code]
  • Yoon Kim, Sam Wiseman, Andrew C. Miller, David Sontag, Alexander M. Rush ICML 2018
• Avoiding Latent Variable Collapse with Generative Skip Models [pdf]
  • Adji B. Dieng, Yoon Kim, Alexander M. Rush, David M. Blei ICML 2018 workshop on Theoretical Foundations and Applications of Deep Generative Models
• Variational Attention for Sequence-to-Sequence Models [pdf][code]
  • Hareesh Bahuleyan, Lili Mou, Olga Vechtomova, Pascal Poupart COLING 2018
• Generating Sentences by Editing Prototypes [pdf][code]
  • Kelvin Guu, Tatsunori B., Hashimoto Yonatan, Oren Percy Liang ACL 2018
• A Hybrid Convolutional Variational Autoencoder for Text Generation [pdf][code]
  • Stanislau Semeniuta, Aliaksei Severyn, Erhardt Barth EMNLP 2017
• Piecewise Latent Variables for Neural Variational Text Processing [pdf]
  • Iulian V. Serban1 and Alexander G. Ororbia II2 and Joelle Pineau3 and Aaron Courville1 EMNLP 2017
• Z-Forcing: Training Stochastic Recurrent Networks [pdf]
  • Anirudh Goyal, Alessandro Sordoni, Marc-Alexandre Côté, Nan Rosemary Ke, Yoshua Bengio NIPS 2017
• Toward Controlled Generation of Text [pdf][code]
  • Zhiting Hu, Zichao Yang, Xiaodan Liang, Ruslan Salakhutdinov, Eric P. Xing ICML 2017
• Improved Variational Autoencoders for Text Modeling using Dilated Convolutions [pdf][code]
  • Zichao Yang, Zhiting Hu, Ruslan Salakhutdinov, Taylor Berg-Kirkpatrick ICML 2017
• Generating Sentences from a Continuous Space [pdf][code]
  • Samuel R. Bowman, Luke Vilnis, Oriol Vinyals, Andrew M. Dai, Rafal Jozefowicz, Samy Bengio CoNLL 2016

Autoencoder based

• Adversarially Regularized Autoencoders for Generating Discrete Structures [pdf][code]
  • Jake Zhao (Junbo), Yoon Kim, Kelly Zhang, Alexander M. Rush, Yann LeCun ICML 2018

Reinforcement learning based

• Long Text Generation via Adversarial Training with Leaked Information[pdf][code]
  • Jiaxian Guo, Sidi Lu, Han Cai, Weinan Zhang, Yong Yu, Jun Wang AAAI 2018
• MaskGAN: Better Text Generation via Filling in the______ [pdf][code]
  • William Fedus, Ian Goodfellow, Andrew M. Dai ICLR 2018
• Adversarial ranking for language generation [pdf][code]
  • Kevin Lin, Dianqi Li, Xiaodong He, Zhengyou Zhang, Ming-Ting Sun AAAI 2018
• Boundary-Seeking Generative Adversarial Networks [pdf][code]
  • R Devon Hjelm, Athul Paul Jacob, Tong Che, Adam Trischler, Kyunghyun Cho, Yoshua Bengio ICLR 2018
• Maximum-Likelihood Augmented Discrete Generative Adversarial Networks (MaliGAN)[pdf]
  • Tong Che, Yanran Li, Ruixiang Zhang, R Devon Hjelm, Wenjie Li, Yangqiu Song, Yoshua Bengio
• SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient [pdf][code (official)][code (non-official)]
  • Lantao Yu, Weinan Zhang, Jun Wang, Yong Yu AAAI 2017

Alternative decode objective

• Learning to Write with Cooperative Discriminators [pdf][code]
  • Ari Holtzman, Jan Buys, Maxwell Forbes, Antoine Bosselut, David Golub, Yejin Choi ACL 2018

Tool and others

• Syntax Maker: an Open Source Natural Language Generation Tool for Finnish [pdf] [code]
  • Mika Hämäläinen, Jack Rueter IWCLUL 2018
• Texar: A Modularized, Versatile, and Extensible Toolbox for Text Generation [pdf][website][github]
  • Zhiting Hu, Zichao Yang, Haoran Shi, Bowen Tan, Tiancheng Zhao,Junxian He, Xiaodan Liang, Wentao Wang, Xingjiang Yu, Di Wang, Lianhui Qin, Xuezhe Ma, Hector Liu, Devendra Singh, Wangrong Zhu, Eric P. Xing ACL 2018
• Texygen: A Benchmarking Platform for Text Generation Models [pdf][code]
  • Yaoming Zhu, Sidi Lu, Lei Zheng, Jiaxian Guo, Weinan Zhang, Jun Wang, Yong Yu SIGIR 2018
• Neural Text Generation: Past, Present and Beyond [pdf]
  • Sidi Lu, Yaoming Zhu, Weinan Zhang, Jun Wang, Yong Yu arxiv 2018

Applications

Stylistic Text (transfer)

• SentiGAN: Generating Sentimental Texts via Mixture Adversarial Networks [pdf]
  • Ke Wang, Xiaojun Wan IJCAI 2018
• Adversarially Regularized Autoencoders for Generating Discrete Structures [pdf][code] (This is AE based ! Put here for the sake of simple hierarchy)
  • Jake Zhao (Junbo), Yoon Kim, Kelly Zhang, Alexander M. Rush, Yann LeCun ICML 2018
• Toward Controlled Generation of Text [pdf][code]
  • Zhiting Hu, Zichao Yang, Xiaodan Liang, Ruslan Salakhutdinov, Eric P. Xing ICML 2017
• Style Transfer Through Back-Translation [pdf]
  • Shrimai Prabhumoye, Yulia Tsvetkov, Ruslan Salakhutdinov, Alan W Black ACL 2018
• Style Transfer in Text: Exploration and Evaluation [pdf]
  • Zhenxin Fu, Xiaoye Tan, Nanyun Peng, Dongyan Zhao, Rui Yan AAAI 2018
• Style Transfer from Non-Parallel Text by Cross-Alignment [pdf]
  • Tianxiao Shen, Tao Lei, Regina Barzilay, Tommi Jaakkola NIPS 2017

(Visual) Dialogue

• Discriminative Deep Dyna-Q: Robust Planning for Dialogue Policy Learning [pdf]
  • Shang-Yu Su Xiujun Li Jianfeng Gao Jingjing Liu Yun-Nung Chen EMNLP 2018
• Deep Dyna-Q: Integrating Planning for Task-Completion Dialogue Policy Learning[pdf]
  • Baolin Peng, Xiujun Li, Jianfeng Gao, Jingjing Liu, Kam-Fai Wong, Shang-Yu Su ACL 2018
• Are You Talking to Me? Reasoned Visual Dialog Generation through Adversarial Learning [pdf]
  • Qi Wu, Peng Wang, Chunhua Shen, Ian Reid, Anton van den Hengel CVPR 2018
• Adversarial Learning for Neural Dialogue Generation [pdf]
  • Jiwei Li, Will Monroe, Tianlin Shi, Sébastien Jean, Alan Ritter, Dan Jurafsky EMNLP 2017
• Learning cooperative visual dialog agents with deep reinforcement learning [pdf]
  • Abhishek Das, Satwik Kottur, José M.F. Moura, Stefan Lee ICCV 2017 , Dhruv Batra1,4
• Polite Dialogue Generation Without Parallel Data [pdf]
  • Tong Niu, Mohit Bansal TACL 2018
• Variational Autoregressive Decoder for Neural Response Generation [[paper]](Coming soon)
  • Jiachen Du, Wenjie Li, Yulan He, Ruifeng Xu, Lidong Bing and Xuan Wang EMNLP 2018
• Unsupervised Discrete Sentence Representation Learning for Interpretable Neural Dialog Generation [pdf]
  • Tiancheng Zhao, Kyusong Lee and Maxine Eskenazi ACL 2018
• Improving Variational Encoder-Decoders in Dialogue Generation [pdf]
  • Xiaoyu Shen, Hui Su, Shuzi Niu and Vera Demberg AAAI 2018
• A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues [pdf]
  • Iulian Vlad Serban, Alessandro Sordoni, Ryan Lowe, Laurent Charlin, Joelle Pineau, Aaron Courville, Yoshua Bengio AAAI 2017
• Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders [pdf]
  • Tiancheng Zhao, Ran Zhao and Maxine Eskenazi ACL 2017
• A Conditional Variational Framework for Dialog Generation [pdf]
  • Xiaoyu Shen, Hui Su, Yanran Li, Wenjie Li, Shuzi Niu, Yang Zhao, Akiko Aizawa, Guoping Long ACL 2017

Image to text

• Recurrent Topic-Transition GAN for Visual Paragraph Generation [pdf]
  • Xiaodan Liang, Zhiting Hu, Hao Zhang, Chuang Gan, Eric P. Xing ICCV 2017
• Towards Diverse and Natural Image Descriptions via a Conditional GAN [pdf]
  • Bo Dai, Sanja Fidler, Raquel Urtasun, Dahua Lin ICCV 2017
• Show, Adapt and Tell: Adversarial Training of Cross-domain Image Captioner [pdf][code]
  • Tseng-Hung Chen, Yuan-Hong Liao, Ching-Yao Chuang, Wan-Ting Hsu, Jianlong Fu, Min Sun ICCV 2017
• Improved Image Captioning via Policy Gradient optimization of SPIDEr [pdf]
  • Liu, Siqi; Zhu, Zhenhai; Ye, Ning; Guadarrama, Sergio; Murphy, Kevin ICCV 2017
• Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training [pdf]
  • Rakshith Shetty, Marcus Rohrbach, Lisa Anne Hendricks, Mario Fritz1 Bernt Schiele ICCV 2017

Other

• Generating Reasonable and Diversified Story Ending Using Sequence to Sequence Model with Adversarial Training [pdf]
  • Zhongyang Li, Xiao Ding and Ting Liu COLING 2018
• Conditional Generative Adversarial Networks for Commonsense Machine Comprehension [pdf]
  • Bingning Wang, Kang Liu, Jun Zhao IJCAI 2017
• Unsuprervised Cipher Cracking Using Discrete GANS [pdf]
  • Aidan N. Gomez, Sicong Huang, Ivan Zhang, Bryan M. Li, Muhammad Osama, Łukasz Kaiser ICLR 2018
• Multimodal Storytelling via Generative Adversarial Imitation Learning [pdf]
  • Zhiqian Chen, Xuchao Zhang, Arnold P. Boedihardjo, Jing Dai, Chang-Tien Lu1, IJCAI 2017
• Compositional Obverter Communication Learning From Raw Visual Input [pdf]
  • Edward Choi, Angeliki Lazaridou, Nando de Freitas ICLR 2018
• Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory [pdf]
  • Hao Zhou, Minlie Huang, Tianyang Zhang, Xiaoyan Zhu, Bing Liu AAAI 2018

Contribution

Please help contribute this list by contacting me or add pull request

Markdown format:

- Paper Name [[pdf]](link) [[code]](link)
  - Author 1, Author 2, Author 3. *Conference'Year*

License

To the extent possible under law, kuanchen has waived all copyright and related or neighboring rights to this work.