論文輪読 2018

このセミナーについて (About this seminar)

担当者が見つけた面白い研究を紹介するセミナーです.以下の内容で構成されます. In this seminar, presenters share interesting studies that they found. The seminar consists of the following elements.

  1. 論文紹介 (Paper presentation): 面白いと思う論文を選んで,その内容を著者の代わりになったつもりで発表する. A presenter reads a paper that he or she finds interesting, and explains the content as if he or she was an author of the paper.
  2. 著者プレゼンの上映と解説 (Watching talk videos): 担当者は面白いと思う研究の発表動画を選び,その論文の内容を把握する.担当者の補足説明を聞きながら,参加者みんなで発表動画を鑑賞する. Finding an interesting video presentation, a presenter reads and understands the content of the paper. Attendees watch the presentation video of the author of the paper. The presenter is expected to explain supplementary information about the research.

前者 (1) はプレゼンテーションの練習を兼ねています.後者 (2) は「よい」プレゼンテーションを鑑賞しながら,英語でのプレゼンテーションやディスカッションに慣れることを狙っています. The former (1) aims at practicing presentations. The latter (2) aims at improving listening and discussion skills in English as we enjoy good presentations.

日時 (Date and time)

  • 16:50- 18:20, Wednesday

参加者 (Atendee)

  • 全員

今後の予定 (Planned schedule)

2019-03-06(Wed) 16:50

  • takase: Ziqiang Cao, Wenjie Li, Sujian Li, Furu Wei. Retrieve, Rerank and Rewrite: Soft Template Based Neural Summarization, ACL 2018. (slide)

  • wan: Ella Rabinovich, Yulia Tsvetkov, Shuly Wintner, Native Language Cognate Effects on Second Language Lexical Choice(video, paper)

過去の記録 (Past seminars)

#1:2018- 04-18(Wed) 16:50

  • Wan: Shuohang Wang, Mo Yu, Jing Jiang, Wei Zhang, Xiaoxiao Guo, Shiyu Chang, Zhiguo Wang, Tim Klinger, Gerald Tesauro, Murray Campbell. Evidence Aggregation for Answer Re-Ranking in Open-Domain Question Answering. to appear in ICLR 2018, 2018. 資料, 論文リンク:ICLR2018
  • Mizuki: Raphael Shu, Hideki Nakayama. Compressing Word Embeddings via Deep Compositional Code Learning, To Appear: ICLR 2018, 2018. 資料, 論文リンク:ICLR(Open Review), 論文リンク:言語処理学会

#2:2018-04-25 (Wed) 15:05

  • Erick: H. Noh, A. Araujo, J. Sim, T. Weyand, B. Han, “Large-Scale Image Retrieval with Attentive Deep Local Features”, Proc. ICCV’17 Materials, Code
  • Okazaki: Melvin Johnson, Mike Schuster, Quoc V. Le, Maxim Krikun, Yonghui Wu, Zhifeng Chen, Nikhil Thorat, Fernanda Viégas, Martin Wattenberg, Greg Corrado, Macduff Hughes, Jeffrey Dean. 2017. Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation. EMNLP 2017. Supplementary materials

#3:2018-05-09 (Wed) 15:05

  • Matsumaru: Maxim Rabinovich, Mitchell Stern, Dan Klein. 2017. Abstract Syntax Networks for Code Generation and Semantic Parsing. ACL 2017. Materials

  • Shimadu: Xuwang Yin, Vicente Ordonez.Obj2Text Generating Visually Descriptive Language from Object Layouts.EMNLP2017.Paper,Talk

#4:2018-05-16 (Wed) 15:05

  • Mizuki: ARORA, Sanjeev; LIANG, Yingyu; MA, Tengyu. A simple but tough-to-beat baseline for sentence embeddings. In: ICLR 2017. (Material )
  • Sasazawa:Nikita Kitaev, Dan Klein.Where is Misty?Interpreting Spatial Descriptors by Modeling Regions in Space.EMNLP 2017(Paper,Talk)

#5:2018-05-23 (Wed) 15:05

  • Shimadu:Srinivasan Iyer, Ioannis Konstas, Alvin Cheung, Luke Zettlemoyer. Summarizing Source Code using a Neural Attention Model. ACL 2016. materials

  • Wan: Snigdha Chaturvedi Haoruo Peng Dan Roth. Story Comprehension for Predicting What Happens Next. EMNLP 2017.paper.Talk

#6:2018-06-06 (Wed) 15:05

  • CANCELED: conflict with JSAI2018
  • Matsumaru: Jiwei Tan, Xiaojun Wan and Jianguo Xiao. 2017. Abstractive Document Summarization with a Graph-Based Attentional Neural Model. ACL 2017. Paper, Talk

#7:2018-06-13 (Wed) 15:05

  • Mizuki: 馬場 雪乃. ヒューマンコンピュテーションとクラウドソーシング. In: 人工知能学会2018, チュートリアル公演(Memo(Ja/En), related topics)
  • Erick Huanyu Yu, Shuo Cheng, Bingbing Ni, Minsi Wang, Jian Zhang, Xiaokang Yang. Fine-grained Video Captioning for Sports Narrative. 2018. To appear in CVPR 2018 material

#8:2018-06-20 (Wed) 15:05

  • Sasazawa: Mark Dredze,Miles Osborne,Prabhanjan Kambadur.Geolocation for Twitter: Timing Matters.NAACL2018 (Material)

  • Mizuki: FELBO, Bjarke, et al. Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm. In: EMNLP 2017. (Memo(Ja), Memo(En))

#9: 2018-06-27 (Wed) 15:05

  • Okazaki: Matthew E. Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, Luke Zettlemoyer. 2018. Deep contextualized word representations. to appear in NAACL 2018. (Material

  • Shimadu: Felix Stahlberg, Bill Byrne. Unfolding and Shrinking Neural Machine Translation Ensembles. to appear in EMNLP 2016 Paper ,Talk

#10:2018-07-04 (Wed) 15:05

  • Yuling: Ivan Habernal, Henning Wachsmuth, Iryna Gurevych, Benno Stein. The Argument Reasoning Comprehension Task: Identification and Reconstruction of Implicit Warrants. To appear in NAACL 2018. (Material )
  • Sasazawa:Yonatan Bisk, Daniel Marcu and Deniz Yuret.Natural Language Communication with Robots.(paper,talk)

#11:2018-07-11 (Wed) 15:05

  • Wan: Sweta Karlekar, Tong Niu, Mohit Bansal. NAACL 2018. Detecting Linguistic Characteristics of Alzheimer’s Dementia by Interpreting Neural Models(material , paper)
  • Matsumaru: Yonatan Belinkov, Nadir Durrani, Fahim Dalvi, Hassan Sajjad, James Glass. ACL 2017. What do Neural Machine Translation Models Learn about Morphology? (paper, talk)

#12:2018-07-18 (Wed) 15:05

  • Shimadu : Zhen Yang , Wei Chen , Feng Wang, Bo Xu. Unsupervised Neural Machine Translation. to appear in ACL2018. material

  • Erick: Xing Niu, Marianna Martindale, Marine Carpua. A Study of Style in Machine Translation: Controlling the Formality of Machine Translation Output. EMNLP 2017 (paper, talk)

#13:2018-07-25 (Wed) 15:05

  • Okazaki: Akira Sasaki, Kazuaki Hanawa, Naoaki Okazaki and Kentaro Inui. Predicting Stances from Social Media Posts using Factorization Machines. to appear in Coling 2018. paper note 1 note 2

  • Mizuki: ZAHEER, Manzil, et al. Deep sets. In: Advances in Neural Information Processing Systems. 2017. p. 3394-3404. (Memo(En))

#14:2018-08-01 (Wed) 15:05

  • Matsumaru: Yen-Chun Chen and Mohit Bansal. Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting. ACL 2018. Material
  • Wan: Materials

#15:2018-09-05 (Wed) 15:05

  • sasazawa:Shuohang Wang,Jing Jiang.MACHINE COMPREHENSION USING MATCH-LSTM AND ANSWER POINTER.ICLR 2017.material paper03_shimadu.pdf (447.2 kB)

  • matsumaru: Elizabeth Clark, Yangfeng Ji, Noah Smith. Neural Text Generation in Stories using Entity Representations as Context. NAACL 2018. Paper, Talk

2018-09-12 Canceled


  • erick: DenseCap: Fully Convolutional Localization Networks for Dense Captioning. Justin Johnson, Andrej Karpathy, Li Fei Fei materials
  • sasazawa:A Question Answering Approach to Emotion Cause Extraction.Lin Gui, Jiannan Hu, Yulan He, Ruifeng Xu, Qin Lu, Jiachen Du.EMNLP 2017.Paper,Talk

#17:2018-09-26 (Wed)

  • Wan: Taiki Miyanishi, Jun-ichiro Hirayama, Atsunori Kanemura, Motoaki Kawanabe. Answering Mixed-Type Questions about Daily Living Episodes. IJCAI 2018.(materials)
  • Shimazu : Longyue Wang, Zhaopeng Tu, Andy Way, Qun Liu. Exploiting Cross-Sentence Context for Neural Machine Translation. EMNLP2017 paper, talk

#18:2018-10-03(Wed) 16:50

  • mizuki: RIBEIRO, Marco Tulio; SINGH, Sameer; GUESTRIN, Carlos. Semantically equivalent adversarial rules for debugging nlp models. In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2018. p. 856-865. material/ja/markdown, material/en/pdf
  • okazaki: Eunsol Choi, Omer Levy, Yejin Choi, Luke Zettlemoyer. Ultra-Fine Entity Typing. ACL 2018, pp. 87-96. paper talk

#19:2018-10-10(Wed) 16:50

  • shimadu : Elena Voita, Pavel Serdyukov, Rico Sennrich, Ivan Titov. Context-Aware Neural Machine Translation Learns Anaphora Resolution. ACL2018. material(pdf) ), material(esa)
  • niwa : Rao, Sudha, and Hal Daumé III. “Learning to Ask Good Questions: Ranking Clarification Questions using Neural Expected Value of Perfect Information.” ACL2018 paper, talk

#20:2018-10-17(Wed) 16:50

  • mizuki(research)20181012_Research_Seminar.pdf
    • alternative to the cancelled presentation
  • yang: No Metrics Are Perfect: Adversarial Reward Learning for Visual StorytellingPaper,Talk

#21:2018-10-19 (Fri) 13:20

  • kawada: Chenyang Huang, Osmar R. Zaiane, Amine Trabelsi, Nouha Dziri. Automatic Dialogue Generation with Expressed Emotions. NAACL 2018. slides paper
  • moriya: Semih Yagcioglu, Aykut Erdem, Erkut Erdem and Nazli Ikizler-Cinbis. RecipeQA: A Challenge Dataset for Multimodal Comprehension of Cooking Recipes. EMNLP 2018. slides paper
  • kanamori: Gongbo Tang, Mathias Müller, Annette Rios, Rico Sennrich. Why Self-Attention? A Targeted Evaluation of Neural Machine Translation Architectures. EMNLP 2018. slides paper

#22:2018-10-24(Wed) 16:50

  • Emanuele: Document Context Neural Machine Translation with Memory Networks. Sameen Maruf and Gholamreza Haffari. ACL2018. [paper] [talk]
  • Erick - BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, Jacob Devlin, Ming-Wei Change, Kenton Lee, Kristina Toutanova Google AI Language (Summary, materials)

#23:2018-10-31(Wed) 16:50

  • matsumaru: Bottom-Up Abstractive Summarization. Sebastian Gehrmann, Yuntian Deng, Alexander M. Rush. EMNLP 2018. material

  • shimadu: Adversarial Feature Matching for Text Generation. Yizhe Zhang, Zhe Gan, Kai Fan, Zhi Chen, Ricardo Henao, Dinghan Shen, Lawre. ICML2017. material(esa)

#24:2018-11-07(Wed) 16:50

  • niwa: Zhao, Tiancheng, and Maxine Eskenazi. “Zero-Shot Dialog Generation with Cross-Domain Latent Actions.” arXiv preprint arXiv:1805.04803 (2018). materials

  • mizuki: LI, Juncen, et al. Delete, Retrieve, Generate: a Simple Approach to Sentiment and Style Transfer. In: NAACL 2018. (material(esa))

#25:2018-11-14(Wed) 16:50

  • takase: Chengyue Gong, Di He, Xu Tan, Tao Qin, Liwei Wang, Tie-Yan Liu. FRAGE: Frequency-Agnostic Word Representation. NIPS 2018. slide

  • wan: Pranav Rajpurkar, Robin Jia, Percy Liang, Know What You Don’t Know: Unanswerable Questions for SQuAD(video,paper)

#26:2018-11-21(Wed) 16:50

  • yang: Barbieri, F., Ballesteros, M., Ronzano, F., & Saggion, H. (2018). Multimodal emoji prediction. Multimodal Emoji Prediction.pdf (24.3 MB)

  • matsumaru: Max Glockner, Vered Shwartz, Yoav Goldberg. Breaking NLI Systems with Sentences that Require Simple Lexical Inferences. ACL 2018. paper, talk

#27:2018-11-30(Fri) 13:20

  • sasazawa:Yanghoon Kim, Hwanhee Lee, Joongbo Shin, Kyomin Jung.(2018).Improving Neural Question Generation using Answer Separation.(material)

  • takase: Chao Jiang, Hsiang-Fu Yu, Cho-Jui Hsieh, Kai-Wei Chang. LearningWord Embeddings for Low-resource Languages by PU Learning. paper, video

    • Actually, the presentation is not good (I think) but the method is interesting.
    • The method is strongly based on the previous study: paper

#28:2018-12-05(Wed) 16:50

  • erick: Knowledgeable Reader: Enhancing Cloze-Style Reading Comprehension with External Commonsense Knowledge video, paper
  • okazaki: Adhiguna Kuncoro, Chris Dyer, John Hale, Dani Yogatama, Stephen Clark, Phil Blunsom. LSTMs Can Learn Syntax-Sensitive Dependencies Well, But Modeling Structure Makes Them Better. ACL 2018. paper, video

#29:2018-12-12(Wed) 16:50

  • mizuki: LIU, Changsheng; HWA, Rebecca. Heuristically Informed Unsupervised Idiom Usage Recognition. In: EMNLP 2018. (material)
  • ema: Triangular Architecture for Rare Language Translation. Shuo Ren, Wenhu Chen, Shujie Liu, Mu Li, Ming Zhou and Shuai Ma. ACL 2018. [paper] [video]

#30:2018-12-19(Wed) 16:50

  • wan: S.Mirzaeia, M.ElYacoubia, S.Garcia-Salicetti, Two-Stage Feature Selection of Voice Parameters for Early Alzheimer’s Disease Prediction.(material)

  • sasazawa:Christopher Clark,Matt Gardner.Simple and Effective Multi-Paragraph Reading Comprehension .ACL 2018.(paper,video)


    canceled because of the winter vacation


canceled because of the winter vacation

2019-01-09(Wed) 16:50

Canceled because of the deadline of The Association for NLP

2019-01-16(Wed) 16:50

Canceled because of the deadline of The Association for NLP

#31: 2019-01-23(Wed) 16:50

  • shimadu: Tu, Zhaopeng Lu, Zhengdong Liu, Yang Liu, Xiaohua Li, Hang. Modeling Coverage for Neural Machine Translation. ACL2016. material

  • yang: Google’s Multilingual Neural Machine Translation System Enabling Zero-Shot Translation-Melvin Johnson, Mike Schuster, Quoc V. Le paper, Talk.

#32: 2019-01-30(Wed) 16:50

  • okazaki: Ian Tenney, Patrick Xia, Berlin Chen, Alex Wang, Adam Poliak, R Thomas McCoy, Najoung Kim, Benjamin Van Durme, Sam Bowman, Dipanjan Das, Ellie Pavlick. What do you learn from context? Probing for sentence structure in contextualized word representations. to appear in ICLR 2019. slides

  • niwa:Xu, Jingjing et al. “A Skeleton-Based Model for Promoting Coherence Among Sentences in Narrative Story Generation.” EMNLP (2018).(Paper, Talk)

#33: 2019-02-06(Wed) 16:50

  • yang: emoji2vec: Learning Emoji Representations from their Description. Eisner, B., Rocktäschel, T., Augenstein, I., Bošnjak, M., & Riedel, S. (2016). slides

  • erick: Learning Visually Grounded Sentence Representations. Douwe Kiela, Alexis Conneau, Allan Jabn, Max Nickel. NAACL 2018 Paper, Talk

2019-02-13(Wed) 16:50

No meeting

#34: 2019-02-20(Wed) 16:50

  • matsumaru: Felix Wu, Angela Fan, Alexei Baevski, Yann Dauphin, Michael Auli. Pay Less Attention with Lightweight and Dynamic Convolutions. To appear in ICLR 2019. slides
  • takase: Jason Lee, Elman Mansimov, Kyunghyun Cho. Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement. EMNLP 2018. pdf, video

#35:2019-02-27(Wed) 16:50

  • niwa:Alnajjar, Khalid, Hadaytullah Kundi, and Hannu Toivonen. “Talent, Skill and Support.” Proceedings of the Ninth International Conference on Computational Creativity, 2018. slide

  • mizuki: FENG, Shi, et al. Pathologies of neural models make interpretations difficult. In: EMNLP 2018. p. 3719-3728. memo