XIANG LI Xiang Li

李 想 Xiang Li
反実機械学習と説明可能なAIに関する研究に従事．
IEEE Student Member．電子情報通信学会 学生員．
E-mail: xiang_li[at]lmd.ist.hokudai.ac.jp

Biography Publication Fellowship and Awards Society Activity

  

Biography

  • 2025/10 ~ Present Hokkaido University, Ph.D. in Information Science
  • 2023/10 ~ 2025/09 Hokkaido University, M.S. in Information Science
  • 2023/04 ~ 2023/09 Hokkaido University, Research Student
  • 2018/09 ~ 2022/06 Shandong University, B.S. in Software Engineering

 

Publication

Journal

  1. 太齊 蓮, 李 想, 五箇 亮太, 斉藤 直輝, 前田 圭介, 鎌田 文幸, 久保 竜志, 川嵜 裕二, 小川 貴弘, 長谷山 美紀, “Multimodal large language modelのIn-context learningによる車載カメラ映像を用いた高速道路附属物および植生の異常検出”, AI・データサイエンス論文集, 2025. [Paper]
  2. Xiang Li, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Enhancing classification models with sophisticated counterfactual images,” IEEE Open Journal of Signal Processing, vol. 6, pp. 89-98, 2025. [Paper]

International Conference

  1. Xiang Li, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Enhanced framework for generating counterfactual images with sophisticated caption and inversion-free image editing,” International Workshop on Advanced Image Technology (IWAIT), Vol. 13510, pp. 88-93, 2025. [Paper]
  2. Xiang Li, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Reinforcing pre-trained models using counterfactual images,” IEEE International Conference on Image Processing (ICIP), pp. 486-492, 2024. [arXiv] [Paper]
  3. Xiang Li, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Improving visual counterfactual explanation models for image classification via CLIP,” IEEE Global Conference on Consumer Electronics (GCCE), pp. 390-391, 2023. [Paper]

Domestic Conference

  1. Xiang Li, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama, “Strategic classification model reinforcement: Introducing language-guided counterfactual images to identify and address classification weaknesses,” Meeting on Image Recognition and Understanding (MIRU), pp. 1-5, 2024.
  2. 李 想, 藤後 廉, 前田 圭介, 小川 貴弘, 長谷山 美紀, “視覚反実仮想機械学習モデルにおける精度向上に関する一検討 – 特徴抽出モデルが精度に与える影響の検証 –,” 令和5年度 電気・情報関係学会北海道支部連合大会, pp. 1-2, 2023.

 

Fellowship and Awards

  1. Hokkaido University Next Generation AI Doctoral Fellowship (2025/10 ~ 2028/09) [Link]
  2. 北海道大学大学院情報科学院 学院長賞 (2025/09)

 

Society Activity

  1. Reviewer, Meeting on Image Recognition and Understanding, 2025 [Link]