Erica Kido Shimomoto

e-mail: kidoshimomoto.e (at) aist.go.jp

I am a researcher at the Artificial Intelligence Research Center (AIRC), National Institute of Advanced Industrial Science and Technology (AIST, member of the Knowledge and Information Research Team).

My research topics involve natural language processing, computer vision and pattern recognition. More specifically, my main line of study is the investigation of subspace based-methods and its applications on word embedding. Recently, I have started to dive into the multimodal world, trying to develop less data-hungry model and/or reduce the computational need of existing models.

I give a brief explanation about my research in the following video: [YouTube]

I love seeing computers doing human jobs really well. However, I am even more interested in understanding how they can do such jobs so well. If you cannot understand it, how can you trust it?

I also enjoy developing applications involving robotics and electronics.

Index

• Publications
• Education
• Work Experience
• Academic events organization
• Grants and Awards
• Invited Talks
• Language Proficiency
• Mechatronics Projects
• Miscellaneous

Publications [top]

Ph.D. Thesis

• Subspace Representation for Natural Language Processing. Department of Computer Science, University of Tsukuba, Doctor of Philosophy in Engineering, 2022.03 (Advisor: Kazuhiro Fukui) [Thesis paper]

Journals

• A Video Representation Based on Slow Feature Analysis for Action Recognition
  Suzana Rita Alves Beleza, Erica K. Shimomoto, Lincon S. Souza, Kazuhiro Fukui
  Machine Learning with Applications, 2023.
• Text classification based on the word subspace representation
  Erica K. Shimomoto, François Portet, Kazuhiro Fukui
  Pattern Analysis and Applications, 2021.
  [pdf] (This is a post-peer-review, pre-copyedit version of an article published in the Journal of Pattern Analysis and Applications. The final authenticated version is available online at http://dx.doi.org/10.1007/s10044-021-00960-6).

International Conferences

• Introducing Spatial Information and a Novel Evaluation Scheme for Open-Domain Live Commentary Generation
  Erica K. Shimomoto*, Edison Marrese-Taylor*, Ichiro Kobayashi, Hiroya Takamura and Yusuke Miyao
  * Authors contributed equally to this work.
  Accepted to Findings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024.
• An empirical study of Definition Modeling with LLMs for the main languages of Latin America
  Erica K. Shimomoto*, Edison Marrese-Taylor*, Enrique Reid
  * Authors contributed equally to this work.
  North American Chapter of the Association for Computational Linguistics Conference: LatinX in AI (LXAI) Research Workshop, 2024.
  [Slides]
• Framework for Measuring the Similarity of Visual and Semantic Structures in Sign Languages
  Matheus Silva de Lima, Ryota Sato, Erica K. Shimomoto, Suzana Rita Alves Beleza, Nobuko Kato and Kazuhiro Fukui
  In Proceedings of the 30th International Workshop on Frontiers of Computer Vision - IW-FCV, 2024.
• Sign language recognition based on subspace representations in the spatio-temporal frequency domain
  Ryota Sato, Suzana Rita Alves Beleza, Erica K. Shimomoto, Matheus Silva de Lima, Nobuko Kato, Kazuhiro Fukui
  In Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - ICPRAM, 2024.
• Towards Parameter-Efficient Integration of Pre-Trained Language Models In Temporal Video Grounding
  Erica K. Shimomoto, Edison Marrese-Taylor, Hiroya Takamura, Ichiro Kobayashi, Hideki Nakayama, and Yusuke Miyao
  In Findings of the 61st Annual Meeting of the Association for Computational Linguistics, 2023.
  [Preprint][Slides][Video][Code]
• A Subspace-Based Analysis of Structured and Unstructured Representations in Image-Text Retrieval
  Erica K. Shimomoto, Edison Marrese-Taylor, Hiroya Takamura, Ichiro Kobayashi, and Yusuke Miyao
  In Proceedings of The Workshop on Unimodal and Multimodal Induction of Linguistic Structures Workshop, December, 2022. [Slides]
• News2meme: An Automatic Content Generator from News Based on Word Subspaces from Text and Image
  Erica K. Shimomoto, Lincon S Souza, Bernardo B. Gatto, Kazuhiro Fukui
  In Proceedings of the International Conference on Machine Vision Applications 2019, Tokyo, Japan, May, 2019.
  [code]
• Text Classification based on Word Subspace with Term-Frequency
  Erica K. Shimomoto, Lincon S. Souza, Bernardo B. Gatto, Kazuhiro Fukui
  In Proceedings of the 2018 International Joint Conference on Neural Networks, Rio de Janeiro, Brazil, July, 2018.
  [Preprint]
• A faster face detection method combining Bayesian and Haar Cascade Classifiers
  Erica K. Shimomoto, Almir Kimura, Ruan Belem
  In Proceedings of the 2015 CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies, Santiago, Chile, October, 2015.
  [pdf]

Domestic Conferences (Japan)

• Zero-shot Definition Modelling for Portuguese
  Edison Marrese-Taylor, Erica K. Shimomoto, Erique Reid
  In 言語処理学会 第30回年次大会 発表論文集, March, 2024.
• Sign language recognition based on subspace representations in the spatio-temporal frequency domain
  Ryota Sato, Suzana Rita Alves Beleza, Erica K. Shimomoto, Matheus Silva de Lima, Nobuko Kato, Kazuhiro Fukui In PRMU2024 講演論文, vol.123, no.409, pp.19-24, March, 2024.
• Subspace representation for text classification with limited training data
  Erica K. Shimomoto, Edison Marrese-Taylor, Hiroya Takamura, Ichiro Kobayashi, and Yusuke Miyao
  In 言語処理学会 第29回年次大会 発表論文集, March, 2023.

Education [top]

• 2019 - 2022: Doctorate degree in Computer Science. University of Tsukuba, Japan, Graduate School of Systems and Information Engineering, Department of Computer Science.
• 2017 - 2019: Master’s degree in Computer Science. University of Tsukuba, Japan, Graduate School of Systems and Information Engineering, Department of Computer Science.
• 2016 - 2017: Research Student. University of Tsukuba, Japan, Graduate School of Systems and Information Engineering, Department of Computer Science.
• 2009 - 2014: Bachelor of Engineering. Amazonas State University, Brazil. Department of Control and Automation Engineering.
• 2012 - 2013: One year undergraduate exchange student. University of California, Berkeley, USA. Electrical Engineering and Computer Science Department.
• 2006 - 2008: Technical course in Mechatronics. Fundação Matias Machline (former Fundação Nokia de Ensino).

Work Experience [top]

• 2024.10 - Now: Researcher at Knowledge and Information Research Team, Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology (AIST).
• 2022.04 - 2024.09: Postdoctoral researcher at Knowledge and Information Research Team, Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology (AIST).
• 2023.12 - 2024.02: Teaching Assistant at the University of Tokyo for the Introduction to Machine Learning class.
• 2022.11 - 2023.01: Teaching Assistant at the University of Tokyo for the Introduction to Machine Learning class.
• 2021.04 - 2022.03: Research Assistant at Knowledge and Information Research Team, Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology (AIST).
• 2021.01 - 2021.03: Technical Staff at Knowledge and Information Research Team, Artificial Intelligence Research Center, National Institute of Advanced Industrial Science and Technology (AIST).
• 2018 - 2020: Teaching Assistant at the University of Tsukuba for the following classes:
  • Computer Science seminar (Master’s, 2020).
  • Image Recognition Engineering (Undergraduate, 2020).
  • Computer Science midterm presentation and seminar (Master’s, 2019).
  • Image Recognition Engineering (Undergraduate, 2018).
• 2018.08 - 2018.10: Research Assistant at the University of Tsukuba, in collaboration with the Center for Artificial Intelligence Research (C-Air) - Japan.
• 2015 - 2016: Network protocol engineer at Samsung Institute of Informatics Amazonia (SIDIA) - Brazil.
• 2015 - 2016: Research Assistant at Innovation Center in Control and Automation (CICARI) - Brazil.
• 2013 - 2014: Research and development intern at Envision Industry of Electronic Products LTDA - Brazil.

Academic events organization [top]

• Organization committee member for the 2021 CollaboTICS, a collaborative workshop among the University of Tsukuba (Japan), University of Grenoble-Alpes (France) and the Ruhr-Universität Bochum (Germany).
• Program chair for the 2020 UGA & UT Workshop, a collaborative workshop between the University of Tsukuba (Japan) and the University of Grenoble-Alpes (France).

Grants and Awards [top]

• Excellent Student Research Paper Award for the paper Framework for Measuring the Similarity of Visual and Semantic Structures in Sign Languages, at the IW-FCV2024.
• Committee Special Award for the paper “Subspace representation for text classification with limited training data”, at the NLP2023, Japan.
• Dean award for research excellence by the University of Tsukuba, 2022, Japan.
• Computer Science Special Award by the University of Tsukuba, for the organization of the 2021 CollaboTICS Workshop, 2022, Japan.
• Computer Science Special Award by the University of Tsukuba, for the organization of the 2020 UGA & UT Workshop, 2021, Japan.
• “Context change detection in diary data for mental health care based on the word subspace representation”, Young Researcher Training Program for the Degree Programs in Systems and Information Engineering (システム情報工学研究科若手研究者育成プログラム), University of Tsukuba, 181000JPY, 2020.9.1-2021.3.31
• Best Poster Award for the paper titled “News2meme: An Automatic Content Generator from News Based on Word Subspaces from Text and Image”, at the IAPR MVA 2019, Japan.
• MEXT Scholarship Awarded by the Japanese Government (2016 - 2022).
• National leader assistant of the Brazilian delegation in the Ship for the World Youth Leaders (SWY27) (2015).
• Brazilian Government Science Without Borders Scholarship to study at the University of California in Berkeley, USA (2012-2013).
• Research program PAIC grantee at the Amazonas State University, Brazil (2009-2012).
• FEBRACE (Brazilian Fair of Science and Engineering) award for the second-best project documentation for the project titled JALLE’S, at 7th FEBRACE, 2009, Brazil.
• Bronze medal at the Amazonas Chemistry Olympiad, Brazil (2008).
• Honorable Mention at the Brazilian Physics Olympiad, Brazil (2008).

Invited Talks [top]

• Invited Talk at JICA, Jordan: AI Research and Development in Japan, AIRC, 2024.09.12.
  Title: Recent Trends on LLMs.
• Invited Talk at Grandes Temas, an on-line talk series organized by the Academia STEM - UEA, 2023.07.05.
  Title: Academic Career abroad.
• Invited Talk at the Recent Trends in Computer Science and Artificial Intelligence (CollaboTICS 2022), 2022.12.16.
  Title: A Subspace-Based Analysis of Structured and Unstructured Representations in Image-Text Retrieval.
• Invited Talk at Code Talk, an on-line talk series organized by the Centro Acadêmico de Ciência da Computação (Computer Science Academic Center) - FCT/UNESP, 2022.12.22
  Title: Academic Career - From undergrad to grad school.
• Invited to the Round table: Engineering and Sciences of nature (Exact Sciences) at the 2021 Study in Japan Latin America Fair - FIEJ LATAM, 2021.02.25.
• Invited Talk at the following webminars of the 2020 Study in Japan Latin America:
  • The Student Life in Japan, 2020.11.4
  • The Scholarships of Japan, 2020.11.5
• Invited Talk at the Knowledge and Information Research Team, Artificial Intelligence Research Center (AIRC), AIST, 2020.09.29.
  Title: Word subspace modeling of text data and its applications.
• Invited Talk at the Rakuten Institute of Technology, 2019.06.28
• Research Talk: News2meme: An Automatic Content Generator from News Based on Word Subspaces from Text and Image \

Language Proficiency [top]

• Portuguese (native)
• JLPT, N2 (2020)
• TOEIC Listening and Reading, 990 (2019)
• TOEFL IBT, 105 (2012)

Mechatronics Projects [top]

I used to work on many mechatronics related projects for classes as well as personal projects. Unfortunately, I haven’t been able to work on new ones since I joined my Master’s, but I would be really glad to work again with mechatronics.

Megumi - Face Recognition System for smartTVs - Envision (TPV), Manaus, Brazil (2013-2014)

• Creators: Erica K. Shimomoto, Ruan Belém (supervisor)
• Face Recognition System for Televisions, developed using C++ and Python.

BoatCar - Amphibious RF Controlled Vehicle - University of California, Berkeley, USA (2012)

• Creators: Erica K. Shimomoto, Jesse Adam Machado, Dennis Manhart, Tony Lei and Xianhan Liu
• 1:5 RF Controlled vehicle that could move from water to ground and vice versa. Final project for the ME102B course of the Unversity of California at Berkeley, Fall 2012.
• Video: YouTube

JigSaw - University of California, Berkeley, USA (2012)

• Creators: Erica K. Shimomoto, Thomas Yun
• Robotic Arm capable of solving jigsaw puzzles, guided by a vision system (control and vision systems developed in Matlab). Final project for the EE125 course of the University of California at Berkeley, Fall 2012.
• Video: YouTube

Virtual Laboratory for Physics Teaching, Amazonas State University, Manaus, Brazil (2009-2012)

• Creators: Erica K. Shimomoto, José Luiz Nunes de Mello (supervisor)
• Virtual Laboratory for Physics Teaching, simulating experiments with gas and collisions, developed using Java language.

JALLE’S - Palletizer Automaton Arm, Fundação Nokia de Ensino, Manaus, Brazil (2008)

• Creators: Erica K. Shimomoto, Adria Brito, Laion Nogueira, Luma Pantoja, Leandro Forte (supervisor)
• Palletizer Automaton Arm. Separates objects into different pallets based on the object color, guided by a vision system (developed in C). Final project for the technical course in Mechatronics at Fundação Nokia de Ensino.
• Video (Warning: Really cringe early 2010 editing!): YouTube

Miscellaneous [top]

Online Content Creation

• YouTube channel S4dako, since 2015
  I upload videos talking about my experience with the MEXT scholarship, and about my life in Japan as an international student and researcher.

Voluntary activities

• Facilitator at the “Toward the realization of dreams through education and career” Workshop, Tsukuba, Japan (2017, 2018 and 2020 editions). Promoted discussion among Brazilian high school students about future education and career options.

Cultural activities

• Regular member at the University of Tsukuba Tokimeki Taiko Juku [Instagram], Tsukuba, Japan (2016-2019, 2021-2023).
• Sub-leader at the Fuugakazan Taiko [Facebook][Instagram] (2009-2012).
  Regular member (2013-2016)
  Manaus, Brazil
• Regular member at the Cal Raijin Taiko [Instagram], Berkeley, USA (2012-2013).
• Regular member at the Fuugakazan Yosakoi Soran, Manaus, Brazil (2010-2012).