Publication

✔︎ Google Scholar, ✔︎ arXiv, ✔︎ SpeakerDeck

Reviewed Journal Papers （論文誌, 英文査読付のみ）

1. Issey Sukeda and Tomonari Sei:
   On the minimum information checkerboard copulas under fixed Kendall’s rank correlation , Statistical Papers, 2024. ✔︎arxiv ✔︎paper

2. Issey Sukeda and Tomonari Sei:
   Frank copula is minimum information copula under fixed Kendall’s τ , Statistics & Probability Letters, 2024. ✔︎arxiv ✔︎paper

3. Issey Sukeda, Masahiro Suzuki, Hiroki Sakaji, Satoshi Kodera:
   Development and analysis of medical instruction-tuning for Japanese large language models , Artificial Intellignece in Health 1(2), 107-116, 2024. ✔︎paper

4. Issey Sukeda, Atsushi Miyauchi, and Akiko Takeda:
   A study on modularity density maximization: Column generation acceleration and computational complexity analysis , European Journal of Operational Research 309, pp. 516–528, 2023. ✔︎paper ✔︎arXiv

5. Shinnosuke Sawano, Satoshi Kodera, Masataka Sato, Susumu Katsushika, Issei Sukeda, … , and Issei Komuro:
   Age prediction from coronary angiography using a deep neural network: Age as a potential label to extract prognosis-related imaging features , Plos one 17(10), 2022. ✔︎paper

Reviewed International Conference Proceedings （国際学会, 査読付のみ）

1. Issey Sukeda:
   Development and bilingual evaluation of Japanese medical large language model within reasonably low computational resources , NeurIPS Workshop Advancements in Medical Foundation Models, 2024.
   ✔︎arxiv ✔︎OpenReview ✔︎GitHub

2. Hiroaki Murakami, Takuya Sasatani, Masanori Sugimoto, Issey Sukeda, Yukiya Mita, Yoshihiro Kawahara:
   SyncEcho: Echo-Based Single Speaker Time Offset Estimation for Time-of-Flight Localization , Proceedings of the 22nd ACM Conference on Embedded Networked Sensor Systems, 2024. ✔︎paper

3. Issey Sukeda, Masahiro Suzuki, Hiroki Sakaji, and Satoshi Kodera:
   JMedLoRA:Medical Domain Adaptation on Japanese Large Language Models using Instruction-tuning , NeurIPS Workshop Deep Generative Models for Health, 2023. ✔︎arxiv ✔︎openreview ✔︎conference ✔︎poster

4. Issey Sukeda, Hiroaki Murakami, Yuki Nishiyama, and Yoshihiro Kawahara:
   Recursive Queueing Estimation Using Smartphone-based Acoustic Ranging , Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems, 2022. ✔︎paper ✔︎conference

5. Shinnosuke Sawano, Satoshi Kodera, Hirotoshi Takeuchi, Issei Sukeda, Susumu Katsushika, and Issei Komuro :
   Masked Autoencoder-Based Self-Supervised Learning for Electrocardiograms to Detect Left Ventricular Systolic Dysfunction , NeurIPS Workshop on Learning from Time Series for Health, 2022. ✔︎poster ✔︎openreview ✔︎conference

6. Yuuki Nishiyama, Hiroaki Murakami, Ryoto Suzuki, Kazusato Oko, Issey Sukeda, Kaoru Sezaki, and Yoshihiro Kawahara:
   MOCHA: Mobile Check-in Application for University Campuses Beyond COVID-19 , Proceedings of the Twenty-Third International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing, 253-258, 2022. ✔︎paper

Preprint （紀要）

1. 峰岸剛基, 高木洋羽, 木澤翔太, 助田一晟, 谷中 瞳 大規模言語モデルにおいて数値属性間で共有されるスケーリングベクトルの解析とその応用 言語処理学会 第31回年次大会発表論文集

2. Issey Sukeda and Takeru Matsuda: Directional statistical graphical modeling of phase-based connectivity ✔︎biorxiv

3. Issey Sukeda and Tomonari Sei:
   Relative local dependence of bivariate copulas ✔︎arxiv

4. Issey Sukeda, Risa Kishikawa, and Satoshi Kodera:
   70B-parameter large language models in Japanese medical question-answering ✔︎arxiv

Projects

LLM

• 日本語医療LLM JMedLLM-7B-v1・・・現在非公開
• 日本語医療モデル評価用ベンチマーク / Japanese Medical Language Model Evaluation Harness
• 日本語医療LLM MedSwallow
• 日本語医療QAデータセット JJSIMQA・・・現在非公開
• 日本語医療LLM JMedLoRA
• 最新LLMモデルまとめ awesome-latest-LLM
• LLama2-70Bの日本語インストラクションチューニングモデル

Statistics

• Regularized score matching of torus graph model applied to EEG phase-based connectivity analyses
• Minimum Information Copula
• OGI (objective general index) implementation in Python

Talks（研究発表）

1. (heading) 65th ISI World Statistics Congress 2025, Hague, Netherlands, October 2025.
   • Poster session
   • Title: Frank copula is the minimum information copula under fixed Kendall’s τ
2. (heading) Hitotsubashi workshop on dependence modeling with applications, Tokyo, Japan, May 2025.
   • Oral presentation
   • Title: Frank copula, minimum information copula, and relative local dependence
3. FDIG 2025, Tokyo, Japan, March 2025.
   • Poster session
   • Title: On the comparison between the minimum information copulas under fixed rank correlations
4. ICSDS 2024, Nice, France, December 2024.
   • Title: Torus graph modelling for EEG analysis
5. RIKEN CBS Retreat 2024, Gotenba, Japan, November 2024.
   • Poster session Poster （1st place in Theory/Tech Award）
   • Title: Torus graph modelling for EEG analysis
6. Time Series, Random Fields and Beyond, Ulm, Germany, September 2024.
   • Title: Torus graph modelling for EEG analysis Press
7. ベイズ研究集会 2024, Sapporo, Japan, September 2024.
   • Title: Regularized score matching of torus graph model applied to EEG phase-based connectivity analyses
8. 統計関連連合大会2024, Tokyo, Japan, September 2024.
   • コンペティションセッション（最優秀報告賞）
   • Title: コピュラのrelative local dependence
9. 統計サマースクール 2024, Yuzawa, Japan, July 2024.
   • Title: Torus graph modelling for EEG analysis
10. 学術変革領域研究（A) 統一理論　第３回領域会議, Wako, Japan, May 2024.
    • Title: Application of Directional Statistics to Brain Data Analysis Poster
11. 言語処理学会第30回年次大会(NLP2024), Kobe, Japan, March 2024.
    • Title: JMedLoRA：Instruction-tuningによる日本語大規模モデルの医療ドメイン適用 Abstract
12. Machine Learning Summer School, Okinawa, Japan, March 2024.
    • Title: On the comparison between the minimum information copulas under fixed Spearman’s ρ and Kendall’s τ
13. NeurIPS Workshop Deep Generative Models for Health, New Orleans, USA, December 2023.
    • Title: JMedLoRA:Medical Domain Adaptation on Japanese Large Language Models using Instruction-tuning
14. 数理科学者と解く！神経科学のオープンプロブレム, Nagano, Japan, November 2023.
    • Title: 「最小情報コピュラとしてのFrankコピュラ」
15. 接合関数（コピュラ）理論の新展開, Tokyo, Japan, September 2023.
    • Title: 「最小情報コピュラとしてのFrankコピュラ」
16. 2023年度 統計関連学会連合大会, Kyoto, Japan, September 2023.
    • セッション：一般講演 多変量解析(3)
    • Title: 最小情報コピュラとしてのFrankコピュラ
17. 64th ISI World Statistics Congress 2023, Ottawa, Canada, July 2023.
    • Contributed Paper Session
    • Title: Minimum information copula under fixed Kendall’s rank correlation Abstract
18. 日本品質学会第118回テクノメトリックス研究会, Tokyo, Japan, June 2023.
    • オンラインで発表.
    • Title: 周波数領域での最適輸送による心電図データ拡張
19. ２０２３年度人工知能学会全国大会（第３７回）, Kumamoto, Japan, June 2023.
    • 一般セッション
    • Title: 周波数領域での最適輸送による心電図データ拡張 Paper
20. 日本応用数理学会第19回研究部会連合発表会 第8回学生研究発表, Okayama, Japan, March 2023.
    • Title: On the comparison on minimun information copulas under fixed Spearman’s ρ and Kendall’s τ
21. 第17回 日本統計学会春季集会, Tokyo, Japan, March 2023.
    • 学生優秀発表賞受賞 会報 報告書
    • Title: On the comparison on minimun information copulas under fixed Spearman’s ρ and Kendall’s τ
22. 統計科学の開拓（科研費シンポジウム）, Kanazawa, Japan, Dec. 2022.
    • 科学研究費・基盤研究（A） 「大規模複雑データの理論と方法論の革新的展開」 （研究代表者：青嶋 誠（筑波大学）, 課題番号：20H00576）報告書
    • Title: Kendallの順位相関係数を固定した下での最小情報コピュラ Abstract
23. ACM SenSys2022, Boston, MA, USA, Nov. 2022.
    • Poster Session
    • Title: Recursive Queueing Estimation Using Smartphone-based Acoustic Ranging
24. International Workshop on Continuous Optimization, online.
    • Title: A study on modularity density maximization: Column generation acceleration and computational complexity analysis arXiv
25. 日本オペレーションズ・リサーチ学会 秋季研究発表会
    • Title: モジュラリティ密度最大化に対する高速な列生成法の設計と計算困難性の解析 Abstract
26. 最適化手法とアルゴリズム ─未来を担う若手研究者の集い 2022─

Misc.

• Preferred Networks Tech Blog
  • Title: 数値シミュレーションデータの低次元潜在空間における時間発展ダイナミクスの学習