Yusuke Mori's Website

Currently in the process of migrating to the following page: https://www.mori.ai

Biography

Yusuke Mori (森 友亮 in Japanese) received Ph.D. in the field of Information Science and Technology from the University of Tokyo, Japan, in 2021.

He completed the doctoral course as a member of Harada-Mukuta Laboratory (Machine Intelligence Laboratory). On the basis of his experience working in neuroscience and brain simulation research in Kanzaki laboratory (belonged from 2011-04 to 2014-03), now he has a strong interest in storytelling, machine learning, and natural language processing, which he believes to have a tight relationship with human creativity.

Keywords

  • Natural Language Processing
  • Natural Language Generation
  • Machine Learning
  • Storytelling
  • Emotions

Education

  • The University of Tokyo — Doctor of Philosophy in the field of Information Science and Technology, 2021

    • Worked on machine learning and natural language processing.
    • Advisor: Tatsuya Harada (Professor)

  • The University of Tokyo — Master of Information Science and Technology, 2014

    • Master’s Thesis: 「昆虫全脳シミュレーションに向けたシナプス分布推定及び実環境との接続手法の提案」（in Japanese）
    • Advisor: Ryohei Kanzaki (Professor)

  • The University of Tokyo — Bachelor of Engineering, 2012

    • Graduation Thesis:「細胞形態を考慮した昆虫一次昆嗅覚中枢シミュレーション」（in Japanese）
    • Advisor: Ryohei Kanzaki (Professor)

Extra Curricular Activities

  • 2021-02 ~ 2021-09: 先端研広報サポーター（Public Relations Supporter at RCAST）
  • 2012-10 ~ 2015-07: GCL student PR Plan Team (founding member)
    • As ​a founding member of public relations planning team of “GCL”​(Graduate Program for Social ICT Global Creative Leaders @ the University of Tokyo), I made the basis of the public relations activity by students.
  • 2010-04 ~ 2012-03: 丁友会委員（Committee member of Teiyukai）
  • 2008-06 ~ 2011-06: Student Journalist of 東京大学新聞社（The University of Tokyo Newspaper: UTNP）
    • From 2008-11 to 2009-10, I led the Science Articles Department as a chief.
    • The editor-in-chief of “Todai 2011: Todai à la carte”, an information book about UTokyo.

Work Experiences

  • 2021-10 ~: Corporate Researcher
  • 2020-01 ~ 2020-01: Research Intern - Gunosy Tech Lab R&D team, Gunosy Inc., Tokyo, Japan
  • 2017-01 ~ 2018-03: Dwango Artificial Intelligence Laboratory, Dwango, Tokyo, Japan
    • 2017-04 ~ 2018-03: Part-time worker
    • 2017-01 ~ 2017-03: Contractor
  • 2012-09 ~ 2012-09: 1-week Intern - Ministry of Economy, Trade and Industry, Tokyo, Japan
  • 2012-08 ~ 2012-09: Intern - Nikkei Business Publications, Inc., Tokyo, Japan

Contact

Papers

Google Scholar Citations

International Journal

  1. Hiroaki Yamane, Yusuke Mori, Tatsuya Harada, “Humor meets morality: Joke generation based on moral judgement,” Information Processing & Management, Volume 58, Issue 3, 2021, 102520, ISSN 0306-4573, https://doi.org/10.1016/j.ipm.2021.102520. (open access)

  2. Yusuke Mori, Hiroaki Yamane, Yoshitaka Ushiku, Tatsuya Harada, “How narratives move your mind: A corpus of shared-character stories for connecting emotional flow and interestingness,” Information Processing & Management, Volume 56, Issue 5, 2019, Pages 1865-1879, ISSN 0306-4573, https://doi.org/10.1016/j.ipm.2019.03.006.

International Conference/Workshop

  1. Yusuke Mori, Hiroaki Yamane, Yusuke Mukuta, Tatsuya Harada, “The Nectar of Missing Position Prediction for Story Completion,” Text2Story 2021 (ECIR 2021, Workshop), 2021. [Video (in the channel of the workshop on YouTube)] [PDF]
  2. Yusuke Mori, Hiroaki Yamane, Yusuke Mukuta, Tatsuya Harada, “Finding and Generating a Missing Part for Story Completion,” 4th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL) (COLING 2020, Workshop), 2020. [PDF (ACL Anthology)] [Code (GitHub)] [Presentation Slide (Workshop website)]
  3. Yusuke Mori, Hiroaki Yamane, Yusuke Mukuta, Tatsuya Harada, “Toward a Better Story End: Collecting Human Evaluation with Reasons,” In Proceedings of the 12th International Conference on Natural Language Generation (INLG 2019), 2019. [PDF (ACL Anthology)]

国内の研究会 (Domestic Workshop)

査読なし (unrefereed), in Japanese

  1. 森 友亮†, 上原康平†, 原田達也, († equal contribution) “視覚・言語融合 Transformer モデルによる画像からの物語文生成,” CAI+CAI first workshop (言語処理学会第 27 回年次大会 ワークショップ), 福岡（オンライン）, 2021 年 3 月. [PDF (Workshop website)] [Presentation Slides (Workshop website)]
    • [English Title] VisualNT-BART: Image to Narrative Generation with Vision and Language Transformer
  2. 後藤 昂彦, 加沢 知毅, 宮本 大輔, ハウプト ステファン 周一, 森 友亮, 神崎 亮平, “大規模 RCGA による単一ニューロンに対する H-H 型モデルパラメータの推定 〜 カイコガ触角葉ニューロンを対象として 〜,” 信学技報, vol. 113, no. 382, NC2013-68, pp. 7-12, 福岡, 2014 年 1 月.
  3. 森 友亮, 加沢 知毅, 宮本 大輔, 神崎 亮平, “昆虫全脳シミューレーションに向けたスーパーコンピュータとのリアルタイム通信の検討,” 信学技報, vol. 113, no. 223, MBE2013-42, NC2013-28, pp. 35-40, 新潟, 2013 年 9 月.

Posters

国内の学会・シンポジウム (Domestic Conference, Symposium)

査読なし (unrefereed), poster

  1. Tomoki Kazawa, Yusuke Mori, Daisuke Miyamoto, Shigehiro Namiki, Hidetoshi Ikeno, Stephan Shuichi Haupt, Ikuko Nishikawa, Ryohei Kanzaki, “Constructing a multi-compartment parallelized simulation of a premotor area of the silkmoth brain,” 日本比較生理生化学会第 35 回大会, 兵庫, 2013 年 7 月.
  2. Daisuke Miyamoto, Tomoki Kazawa, Stephan Shuichi Haupt, Shigehiro Namiki, Yohei Sato, Hidehito Sato, Yusuke Mori, Ryota Kobayashi, Hidetoshi Ikeno, Ikuko Nishikawa, Ryohei Kanzaki, “IOSSIM: the framework for constructing multi-compartment neuron models and large scale simulations,” 日本比較生理生化学会第 34 回大会, 神奈川, 2012 年 7 月.

査読なし (unrefereed), poster, in Japanese

  1. 加沢 知毅, 宮本 大輔, 森 友亮, ハウプト ステファン 周一, 並木 重宏, 神崎 亮平, 池野 英利, 五十嵐 吉輝, 山岸 嘉彦, 小野島 隆之, 小杉 展弘, 西川 郁子, “IOSSIM（昆虫全脳シミュレータ）の開発と応用,” 次世代生命体統合シミュレーションソフトウェアの研究開発（ライフ）公開シンポジウム, 東京, 2013 年 3 月.

Projects

Missing Position Prediction

We proposed a new task of story understanding “Missing Position Prediction”, aiming human creative writing assistance [Mori et al., LaTeCH-CLfL 2020, Text2Story 2021].

You can access the code of the prototype demonstration system from here.

The code can be run in your local environment or the Google Colab.

mppsc-demo-prototype

Other Activities

Open Source Software commitment

spacy-course Japanese translation project

  • Participated in a project to enable users to use Japanese models in Japanese translated course.

Hugging Face Datasets Sprint 2020

  • The detail of the sprint is shown in HF newsletter.
  • As one of the 275+ participants, I joined the open-source community effort and create PRs of adding datasets.
  • The datasets added in the sprint is released in Hugging Face Datasets v1.2.

Presentation at Study Group

第七回　全日本コンピュータビジョン勉強会　 CVPR2021読み会

第六回　全日本コンピュータビジョン勉強会　 Transformer 論文読み会

I introduced “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale” [Dosovitskiy et al, ICLR 2021] at the 6th japancv, on 2021-04-18.

第 41 回　コンピュータビジョン勉強会＠関東

I introduced “A Hierarchical Approach for Generating Descriptive Image Paragraphs” [Krause et al., CVPR 2017] at the 41st kantocv, on 2017-08-19.

Qualifications

Information Science and Technology

  • Passed Applied Information Technology Engineer Examination (AP) (2015-06)
  • Passed Fundamental Information Technology Engineer Examination (FE) (2012-11)

Language

  • TOEIC L&R 915 (2018-09)

Others

  • 3rd grade Certified Specialist of Intellectual Property Management（administration） (2015-04)