Kosuke Yamada (山田 康輔)
Research Scientist at CyberAgent, Inc (AI Lab)
Collaborative Researcher at Sasano Group, Nagoya University, Japan
Ph.D (Informatics)
I am from Takeda-Sasano Group, Nagoya University, Japan.
I major in Natural Language Processing.
I'm interested in frame semantics, question answering, summarization, and discourse analysis.
Email: yamada.kosuke.v1@s.mail.nagoya-u.ac.jp
Resume: Kosuke-Yamada
Google Schalar: Kosuke Yamada
X: @kosyamada
LinkedIn: kosuke-yamada
GitHub: https://github.com/Kosuke-Yamada
Atcoder: Kosyamada
EducationPublication
2021/4~2024/3
Department of Intelligent Systems, Graduate School of Informatics, Nagoya University
Doctor of Philosophy (Informatics)Ph.D. Candidate
2019/4~2021/3
Department of Intelligent Systems, Graduate School of Informatics, Nagoya University
Master of Informatics
2015/4~2019/3
Department of Electrical Engineering Electronics, and Information Engineering, School of Engineering, Nagoya University
Bachelor of Engineering
2023/9~2023/11 (3 months)
Natural Language Processing Research Group, Department of Computer Science, University of Liverpool
Visiting Research StudentPh.D. Candidate
Publication
Book
Gijutsu-Hyohron Co., Ltd. (2023/7)
Introduction to Large Language Models (in Japanese) [link] [amazon]
Ikuya Yamada, Masatoshi Suzuki, Kosuke Yamada, Ryokan Ri
Journal
Journal of Information Processing, Information Processing Society of Japan, Vol.X, No.X, pp.X-X, 2024
Acquiring Frame Element Knowledge with Deep Metric Learning for Semantic Frame Induction (in Japanese) [to appear]
Kosuke Yamada, Ryohei Sasano, Koichi Takeda
Journal of Natural Language Processing, The Association for Natural Language Processing, Vol.30, No.4, pp.1130-1150, 2023
Semantic Frame Induction with Deep Metric Learning (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, Koichi Takeda
Journal of Natural Language Processing, The Association for Natural Language Processing, Vol.29, No.2, pp.395-415, 2022 Best Paper Award
Semantic Frame Induction using Masked Word Embeddings and Two-Step Clustering (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, Koichi Takeda
Transactions of the Japanese Society for Artificial Intelligence: AI, The Japanese Society for Artificial Intelligence, Vol.35, No.4, pp.B-K22_1-12, 2020
Incorporating Textual Information on User Behavior for Personality Prediction (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, Koichi Takeda
International Conference
EMNLP 2023 (2023/12)
Transformer-based Live Update Generation for Soccer Matches from Tweets [link]
Masashi Oshika, Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
ACL SRW 2023 (2023/7)
Building a Buzzer-quiz Answering System [link]
Naoya Sugiura, Kosuke Yamada, Ryohei Sasano, Koichi Takeda, and Katsuhiko Toyama
ACL 2023 Findings (2023/7)
Acquiring Frame Element Knowledge with Deep Metric Learning for Semantic Frame Induction [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
EACL 2023 (2023/5)
Semantic Frame Induction with Deep Metric Learning [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
EMNLP 2021 (2021/11)
Transformer-based Lexically Constrained Headline Generation [link]
Kosuke Yamada, Yuta Hitomi, Hideaki Tamori, Ryohei Sasano, Naoaki Okazaki, Kentaro Inui, and Koichi Takeda
ACL-IJCNLP 2021 (2021/8)
Semantic Frame Induction using Masked Word Embeddings and Two-Step Clustering [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
ACL-IJCNLP 2021 Findings (2021/8)
Verb Sense Clustering using Contextualized Word Representations for Semantic Frame Induction [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
EMNLP 2020 Findings (2020/11)
Sequential Span Classification with Neural Semi-Markov CRFs for Biomedical Abstracts [link]
Kosuke Yamada, Tsutomu Hirao, Ryohei Sasano, Koichi Takeda, and Masaaki Nagata
ACL 2019 Student Research Workshop (2019/7)
Incorporating Textual Information on User Behavior for Personality Prediction [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
Domestic Conference
The 37th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI) (2023/6) JSAI Annual Conference Award
Generating live sports updates from Twitter using T5 (in Japanese) [link]
Masashi Oshika, Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
The 37th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI) (2023/6)
Acquiring Frame Element Knowledge with Deep Metric Learning for Semantic Frame Induction (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
29th Annual Meeting of the Language Processing Society of Japan (NLP) (2023/3)
Constructing Buzzer Quiz Answering System and Estimating Correct Answer Rate in Process of Reading Quiz (in Japanese) [link]
Naoya Sugiura, Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
29th Annual Meeting of the Language Processing Society of Japan (NLP) (2023/3)
Semantic Frame Induction of Japanese Verb with Deep Metric Learning (in Japanese) [link]
Chihiro Yano, Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
29th Annual Meeting of the Language Processing Society of Japan (NLP) (2023/3)
Semantic Frame Induction with Deep Metric Learning (in Japanese) [link] [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
27th Annual Meeting of the Language Processing Society of Japan (NLP) (2021/3) Committee Special Award, Sponsor Award
Headline Generation including Given Phrases (in Japanese) [link]
Kosuke Yamada, Yuta Hitomi, Hideaki Tamori, Naoaki Okazaki, and Kentaro Inui
27th Annual Meeting of the Language Processing Society of Japan (NLP) (2021/3)
Semantic Frame Induction using Masked Word Embeddings and Two-Step Clustering (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
244th meeting of Special Interest Group of Natural Language Processing (IPSJ-SIGNL) (2020/7)
Investigating the Usefulness of Contextualized Word Representations in Semantic Frame Induction (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
26th Annual Meeting of the Language Processing Society of Japan (NLP) (2020/3)
Structural Analysis with Neural Semi-Markov CRFs for Abstracts (in Japanese) [link]
Kosuke Yamada, Tsutomu Hirao, Ryohei Sasano, Koichi Takeda, and Masaaki Nagata
25th Annual Meeting of the Language Processing Society of Japan (NLP) (2019/3)
User Personality Prediction Using Similarity Behavioral Properties in Social Media (in Japanese) [link]
Kosuke Yamada, Ryohei Sasano, and Koichi Takeda
Others
Journal of Natural Language Processing, The Association for Natural Language Processing, 2023 (2023/6)
Explanation of "Semantic Frame Induction using Masked Word Embeddings and Two-Step Clustering" (in Japanese) [link]
Kosuke Yamada
Journal of Natural Language Processing, The Association for Natural Language Processing, Vol.28, No.4, pp.1325-1330, 2021 (2021/12)
Semantic Frame Induction of Verbs using Contextualized Word Embedding (in Japanese) [link]
Kosuke Yamada
Work Experience
Research
2022/7 ~ 2023/3
Studio Ousia Inc.
Research Intern
2022/4 ~ 2024/3
Japan Society for the Promotion of Science (JSPS)
Research Fellow (DC2)
2020/5 ~ 2024/3
RIKEN AIP (Language Information Access Technology Team)
Research Part-time Worker
2020/4 ~2021/11, 1 year 8 months
The Asahi Shimbun Company
Research Intern
2019/8, 1 month
NTT Communication Science Laboratories
Research Intern
Engineer
2019/4 ~ 2019/10, 7 months
TRYETING Inc.
Research Development Engineering Intern
2019/9, 2 weeks
SoftBank Corporation
Machine Learning Engineer Intern
2018/8, 2 weeks
Yahoo! JAPAN Corporation
Software Engineer Intern
Assistant
2022/10 ~ 2023/3, 6 months
Data Science Exercise B at Nagoya University [syllabus] [syllabus]
Qualified Teaching Assistant
2019/4 ~ 2019/8, 6 months
Innovation Practice Course for Comprehensive Engineering at Nagoya University [link]
Teaching Assistant
Activity
Volunteer
2022/4 ~ 2024/3
Young Researcher Association for NLP Studies Committee (YANS) [link]
Talk
3rd LLM-jp Study Meeting (2023/7)
Introducing the paper "Benchmarking Large Language Model Capabilities for Conditional Generation" (in Japanese) [link]
Ai Robot Science Symposium 2023 (2023/7)
Automatic Construction of Semantic Frame Knowledge as Foundational World Knowledge (in Japanese)
14th State-of-the-art NLP Study Meeting (2022/9)
Introducing the paper "A Contrastive Framework for Learning Sentence Representations from Pairwise and Triple-wise Perspective in Angular Space" (in Japanese) [link]
23rd Meeting at NLP Study Group for Doctoral Students (2022/7)
Toward Constructing High-Quality Semantic Frame Knowledge of Verbs (in Japanese) [link]
13th State-of-the-art NLP Study Meeting (2021/9)
Introducing the paper "ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfer" (in Japanese) [link]
12th State-of-the-art NLP Study Meeting (2020/9)
Introducing the paper "BERTRAM: Improved Word Embeddings Have Big Impact on Contextualized Model Performance" (in Japanese) [link]
Academia Meets Amazon (2019/12)
Incorporating Textual Information on User Behavior for Personality Prediction (in English)
Program
2020/2, 1 week
NC State Executive Education @ North Carolina State University in USA [program] [certificate]
Hackathon
2021/8, 2 weeks
Young Researcher Association for NLP Studies Committee (YANS) [project]
2020/2, 2 days
The Asahi Shimbun Company [project] [blog]
2019/9, 1 day
PwC Consulting LLC [project]
2019/5, 2 days
Gunocy Inc. [project]
2019/3, 5 days
Recruit Technologies Co., Ltd. [project] [blog]
Others
2015/4 ~ 2018/3, 3 years
GAMUSHARA seisyun oendan [YouTube]
2015/7, 1 week
Mississauga Friendship Association [HP]
Award
Research
2023/6
JSAI Annual Conference Award at the 37th Annual Conference of the Japanese Society for Artificial Intelligence [link]
2023/3
Best Paper Award at the Association for Natural Language Processing in 2022 [link]
2022/4
Student's Choice Award for Outstanding Presentation at Nagoya University Doctoral Program Midterm Presentation
2022/4
Teacher's Choice Award for Outstanding Presentation at Nagoya University Doctoral Program Midterm Presentation
2021/3
Committee Special Award at the 27th Annual Meeting of the Language Processing Society of Japan (NLP) [link]
2021/3
Sponsor Award at the 27th Annual Meeting of the Language Processing Society of Japan (NLP) [link]
Business
2018/12
Idea Award at the 2018 Student Business Plan Contest [link]
Qualification
2023/01
Duolingo: 110 (Literacy: 125, Comprehension: 115, Conversation: 80, Production: 80) [link]
2022/03
IELTS: 6.0 (Listening: 6.0, Reading: 6.5, Speaking: 5.0, Writing: 5.5) [link]
2021/07
TOEIC: 810 (Listening: 405, Reading: 405) [link]
2018/12
Information Security Management [link]
2018/6
Database Specialist [link]
2017/12
Applied Information Technology Engineer [link]
2017/6
Fundamental Information Technology Engineer [link]
Scholarship
2022/4 ~ 2024/3, 2 years
Japan Society for Promotion of Science (JSPS) Research Fellowship for Young Scientists (DC2) [link]
200,000 yen / month
2021/4 ~ 2022/3, 1 year
Japan Student Services Organization (JASSO) First-Class Scholarship Repayment Waiver Recipient [link]
122,000 yen / month
2021/4 ~ 2022/3, 1 year
Nagoya University Interdisciplinary Frontier Fellowship [link]
180,000 yen / month
2019/4 ~ 2020/3, 1 year
JEES/Softbank AI Human Resource Development Scholarship [link]
1,000,000 yen / year