publications

2026

1. ICPR

   Tunable MAGMAX: Preference-Aware Model Merging for Continual Learning

   K. Hiroshima, K. Uchida, and S. Shirakawa

   In Proceedings of the International Conference on Pattern Recognition (ICPR)to appear , 2026

   Bib

   @inproceedings{hiroshima2026tunable,
     title = {Tunable {MAGMAX}: Preference-Aware Model Merging for Continual Learning},
     author = {Hiroshima, K. and Uchida, K. and Shirakawa, S.},
     booktitle = {Proceedings of the International Conference on Pattern Recognition (ICPR)},
     address = {Lyon, France},
     year = {2026},
   }

2. NeurIPS

   BBOWP-Bench: Evaluating LLMs on Black-Box Optimization Word Problems

   Y. Yamada, K. Hiroshima, N. Yoshinari, and 2 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), Evaluation and Datasets Trackto appear , 2026

   arXiv Bib Code

   @inproceedings{yamada2026bbowp,
     title = {{BBOWP-Bench}: Evaluating {LLMs} on Black-Box Optimization Word Problems},
     author = {Yamada, Y. and Hiroshima, K. and Yoshinari, N. and Uchida, K. and Shirakawa, S.},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS), Evaluation and Datasets Track},
     address = {Sydney, Australia},
     year = {2026},
   }

3. FAN

   Evaluating Large Language Models on Black-Box Optimization Word Problems

   Y. Yamada, K. Hiroshima, N. Yoshinari, and 2 more authors

   In The 34th Intelligent System Symposium (FAN2026), 2026

   Best Paper Award arXiv Bib Code

   FAN2026 Best Paper Award

   @inproceedings{yamada2026fan,
     title = {Black-Box Optimization Word Problems における大規模言語モデルの評価},
     author = {Yamada, Y. and Hiroshima, K. and Yoshinari, N. and Uchida, K. and Shirakawa, S.},
     booktitle = {第34回インテリジェント・システム・シンポジウム (FAN2026)},
     address = {京都, Japan},
     year = {2026},
   }

4. JSAI

   Autoformulation Benchmark Suite of Black-Box Optimization Problems Described in Natural Language

   Y. Yamada, K. Hiroshima, N. Yoshinari, and 2 more authors

   In The 40th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI2026), 2026

   Bib

   @inproceedings{yamada2026jsai,
     title = {ブラックボックス最適化問題を説明する自然言語記述を入力とした自動定式化ベンチマークの構築},
     author = {Yamada, Y. and Hiroshima, K. and Yoshinari, N. and Uchida, K. and Shirakawa, S.},
     booktitle = {2026年度人工知能学会全国大会 (JSAI2026)},
     address = {群馬, Japan},
     year = {2026},
   }

2025

1. ESC

   Enhancing Atherosclerosis Risk Prediction with Strategic Feature and Case Selections in Large Language Model

   H. Murase, K. Hiroshima, K. Uchida, and 6 more authors

   In ESC Congress 2025, 2025

   Bib

   @inproceedings{murase2025esc,
     title = {Enhancing Atherosclerosis Risk Prediction with Strategic Feature and Case Selections in Large Language Model},
     author = {Murase, H. and Hiroshima, K. and Uchida, K. and Ohashi, M. and Kashihara, N. and Viera, A. J. and Daida, H. and Shirakawa, S. and Yano, Y.},
     booktitle = {ESC Congress 2025},
     address = {Madrid, Spain},
     year = {2025},
   }

2. SSI

   Controlling Forgetting via Model Merging in Continual Learning

   K. Hiroshima, K. Uchida, and S. Shirakawa

   In The 39th Annual Conference of SSI2025, 2025

   SSI Excellent Presentation Award Bib

   SSI Excellent Presentation Award

   @inproceedings{hiroshima2025ssi,
     title = {モデルマージを利用した継続学習アルゴリズムにおける忘却度調整法の検討},
     author = {Hiroshima, K. and Uchida, K. and Shirakawa, S.},
     booktitle = {第39回計測自動制御学会 システム・情報部門 学術講演会 (SSI2025)},
     address = {岩手, Japan},
     year = {2025},
   }