Nanato Izumi, Toshinori Yamauchi, Hiroshi Kera, and Kazuhiko Kawamoto
SHAPIN-CLIP: Game-Theoretic Explanation of CLIP Image-Text Matching
Proceedings of International Symposium on Nonlinear Theory and Its Applications, 2026
Publications
Publications are listed by category and sorted by year.
Journals & Conferences (Peer-Reviewed)
2026
Sai Ganesh Nagarajan, Toshinori Yamauchi, and Hiroshi Kera
Seeing Through Distractions: Stable Attribution via the Core
International Conference on Machine Learning (ICML) Workshop, OpenReview, 2026
Yuhei Senuma, Toshinori Yamauchi, Tomoyuki Okuda, Hideaki Sasazawa, and Yasutaka Toyoda
LLM-powered fab operation assistant for CD-SEM recipe creation
Metrology, Inspection, and Process Control XL, Proceedings of SPIE, Vol. 13981, 139814N, 2026
Toshinori Yamauchi
Spatial Sensitive Grad-CAM++: Towards High-Quality Visual Explanations for Object Detectors via Weighted Combination of Gradient Maps
Computer Vision and Image Understanding (CVIU), vol. 264, pp. 104658, 2026
2025
Toshinori Yamauchi
Spatial Sensitive Grad-CAM: Toward Instance-Specific Explanations for Object Detectors by Incorporating Spatial Sensitivity
IEEE Access, vol. 13, pp. 202086-202102, 2025
2024
Tomoyuki Okuda, Toshinori Yamauchi, San Kang, Yoongyu Park, Kiwoong Lee, Ilyong Lee, Hyewon Park, Gwangseob Lim, Youjoung Jun, Yoonyoung Jun, Gyeongyun Shin
High throughput CD-SEM metrology using image denoising based on deep learning
Metrology, Inspection, and Process Control XXXVIII, Proceedings of SPIE, Vol. 12955, 129551Z, 2024
Toshinori Yamauchi
Spatial Sensitive Grad-CAM++: Improved Visual Explanation for Object Detectors via Weighted Combination of Gradient Map
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pp. 8164-8168, 2024
OralSelected as Spotlight Award
2023
Toshinori Yamauchi, Kentaro Ohira, Takefumi Kakinuma
Positive/Negative Decision via Outlier Detection Toward Automatic Performance Evaluation for Defect Detector
IEEE Transactions on Semiconductor Manufacturing, vol. 36, pp. 527-536, 2023
2022
Toshinori Yamauchi, Kentaro Ohira, Takefumi Kakinuma
Positive/Negative Decision via Outlier Detection Toward Automatic Performance Evaluation for Defect Detector
International Symposium on Semiconductor Manufacturing (ISSM), 2022
OralBest Paper Award
Toshinori Yamauchi, Masayoshi Ishikawa
Spatial Sensitive GRAD-CAM: Visual Explanations for Object Detection by Incorporating Spatial Sensitivity
IEEE International Conference on Image Processing (ICIP), pp. 256-260, 2022
Oral
Preprints
2026
Toshinori Yamauchi, Hiroshi Kera, Kazuhiko Kawamoto
Zero-Shot Faithful Textual Explanations via Directional-Derivative Influence on Predictions
arXiv:2605.16877, 2026
2025
Conferences and Others (Non-Peer-Reviewed)
2026
佐藤 拓斗, 山内 俊典, 計良 宥志, 川本 一彦
視覚言語モデルの命名困難な特徴に対する属性ヒントを用いた自動命名
画像の認識・理解シンポジウム, 2026
Oral (Acceptance Rate: 33%)
Toshinori Yamauchi, Hiroshi Kera, Kazuhiko Kawamoto
Zero-Shot Textual Explanations via Directional-Derivative Influence on Predictions
画像の認識・理解シンポジウム, 2026
Oral (Acceptance Rate: 33%)
Nanato Izumi, Toshinori Yamauchi, Hiroshi Kera, Kazuhiko Kawamoto
SHAPIN-CLIP: Game-Theoretic Explanation of CLIP Image-Text Matching
画像の認識・理解シンポジウム, 2026
2025
Toshinori Yamauchi, Hiroshi Kera, Kazuhiko Kawamoto
Explaining Object Detectors via Collective Contribution of Pixels
画像の認識・理解シンポジウム MIRU, 2025
泉七翔, 山内俊典, 計良宥志, 川本一彦
視覚言語モデルにおける画像―テキスト間対応のゲーム理論的可視化
画像の認識・理解シンポジウム MIRU, 2025
2024
山内俊典, 計良宥志, 川本一彦
ピクセル間の相互作用値を考慮した物体検出モデルに対する視覚的説明手法
画像の認識・理解シンポジウム MIRU, 2024
2023
山内俊典
Spatial Sensitive Grad-CAM++: 勾配の重みづけ平均考慮による物体検出モデルに対する視覚的説明手法の改良
画像の認識・理解シンポジウム MIRU, 2023
Talks
2026
Toshinori Yamauchi
Explaining Object Detectors via Collective Contribution of Pixels
画像の認識・理解シンポジウム, 2026
Invited Talk
Patents
2025
Toshinori Yamauchi, Masayoshi Ishikawa, Takefumi Kakinuma, Masaki Hasegawa, Kentaro Ohira, Yasutaka Toyoda
Image recognition system
US12423941B2, 2025
Published: 2025-09-23
Toshinori Yamauchi, Yasuhiro Yoshida, Masayoshi Ishikawa, Takefumi Kakinuma, Masaki Hasegawa, Kentaro Ohira, Yasutaka Toyoda
Labeled training data creation assistance device and labeled training data creation assistance method
US20250054270A1, 2025
Published: 2025-02-13
2023
Yasuhiro Yoshida, Masayoshi Ishikawa, Toshinori Yamauchi, Kosuke Fukuda, Hiroyuki Shindo
Defect inspection apparatus
US20230402249A1, 2023
Published: 2023-12-14
Awards & Honors
2026
Audience Award, Meeting on Image Recognition and Understanding (MIRU 2026) — “Zero-Shot Faithful Textual Explanations via Directional-Derivative Influence on Predictions.”
Toshinori Yamauchi, Hiroshi Kera, Kazuhiko Kawamoto
Student Encouragement Award, Meeting on Image Recognition and Understanding (MIRU 2026) — “Zero-Shot Faithful Textual Explanations via Directional-Derivative Influence on Predictions.”
Toshinori Yamauchi
Audience Award, Meeting on Image Recognition and Understanding (MIRU 2026) — 「視覚言語モデルの命名困難な特徴に対する属性ヒントを用いた自動命名」
佐藤 拓斗, 山内 俊典, 計良 宥志, 川本 一彦
Selected as a Highlight Paper, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) — “Explaining Object Detectors via Collective Contribution of Pixels.”
Toshinori Yamauchi, Hiroshi Kera, Kazuhiko Kawamoto
2024
Spotlight Award, IEEE/CVF CVPR Workshops — “Spatial Sensitive Grad-CAM++: Improved Visual Explanation for Object Detectors via Weighted Combination of Gradient Map.”
Toshinori Yamauchi
2022
Best Paper Award, International Symposium on Semiconductor Manufacturing (ISSM) — “Positive/Negative Decision via Outlier Detection Toward Automatic Performance Evaluation for Defect Detector.”
Toshinori Yamauchi, Kentaro Ohira, Takefumi Kakinuma
Grants / Fellowships
2026
Selected for the FY2026 Supporting Pioneering Research through AI for 1,000 Discovery challenges (SPReAD)
Selected for the Chiba University Open Recruitment for International Exchange Program: Overseas Travel Support Program for Graduate Students