Sho Yokoi (横井 祥)

alt text 

Sho Yokoi

Research Interest

  • Geometry of word embedding space

Publications

[Google Scholar] [dblp] [Semantic Scholar] [ACL Anthology] [OpenReview] [Scopus] [researchmap]

Preprint

  • Yoichi Ishibashi, Sho Yokoi, Katsuhito Sudoh, Satoshi Nakamura.
    Beyond Vectors: Subspace Representations for Set Operations of Embeddings.
    [arXiv] [code]

2024

2023

  • Momose Oyama, Sho Yokoi, Hidetoshi Shimodaira.
    Norm of Word Embedding Encodes Information Gain.
    In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp.2108–2130, 2023.
    [arXiv] [paper] [OpenReveiw]

  • Hiroaki Yamagiwa, Sho Yokoi, Hidetoshi Shimodaira.
    Improving word mover's distance by leveraging self-attention matrix.
    In Findings of the Association for Computational Linguistics: EMNLP 2023, pp.11160–11183, 2023.
    [arXiv] [paper] [OpenReview] [code]

  • Hiroto Kurita, Goro Kobayashi, Sho Yokoi, Kentaro Inui.
    Contrastive Learning-based Sentence Encoders Implicitly Weight Informative Words.
    In Findings of the Association for Computational Linguistics: EMNLP 2023, pp.10932–10947, 2023.
    [arXiv] [paper] [OpenReview] [code]

  • Yuki Arase, Han Bao, Sho Yokoi.
    Unbalanced Optimal Transport for Unbalanced Word Alignment.
    In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL), pp.3966–3986, 2023.
    [arXiv] [paper] [code]

  • Goro Kobayashi, Tatsuki Kuribayashi, Sho Yokoi, Kentaro Inui.
    Transformer Language Models Handle Word Frequency in Prediction Head.
    In Findings of the Association for Computational Linguistics: ACL 2023, pp.4523–4535, 2023.
    [arXiv] [paper] [code]

  • Masaki Sakata, Sho Yokoi, Benjamin Heinzerling, Kentaro Inui.
    Geometric Locality of Entity Embeddings in Masked Language Models.
    The 61st Annual Meeting of the Association for Computational Linguistics: Student Research Workshop, 2023.

2022

2021

  • Ayato Toyokuni, Sho Yokoi, Hisashi Kashima, Makoto Yamada.
    Computationally Efficient Wasserstein Loss for Structured Labels.
    In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Student Research Workshop, pp.1–7, 2021.
    [arXiv] [paper]

2020

2018

2017

2016

Publications (Domestic)

Non-refereed publications, Talks

Domestic Conferences, Workshops

Education

  • Ph.D. in Information Sciences, April 2017 – March 2020. Tohoku University. Supervisor: Prof. Kentaro Inui.

  • M.S. in Information Sciences, April 2015 – March 2017. Tohoku University. Supervisor: Prof. Kentaro Inui.

  • B.S. in Engineering, April 2011 – March 2015. Kyoto University. Supervisor: Prof. Hisashi Kashima.

Professional Activities

International Conference (reviewer, chair)

Domestic Conference (organizer)

Others (organizer)

Member

Teaching

  • Experiment C (Neural NLP), Tohoku University, Fall 2023.

  • Programming Practice A (Algorithm), Tohoku University, Spring 2023.

  • Programming Practice A (Algorithm), Tohoku University, Spring 2022.

  • Programming Practice A (Algorithm), Tohoku University, Spring 2021.

  • Programming Practice A (Algorithm), Tohoku University, Spring 2020.