🎉 ICLR 2025 Accepted 🎉
See you in Singapore!
Rethinking Invariance Regularization in Adversarial Training to Improve Robustness-Accuracy Trade-off
https://t.co/0UTynUrh5m
#ICLR2025
🎉 Thrilled to announce that our paper "Text-Printed Image: Bridging the Image-Text Modality Gap for Text-centric Training of Large Vision-Language Models" received the CVPR Compute Gold Star at #CVPR2026!
Congrats to our co-authors!!🙌
https://t.co/mCGCXz2xrg
本日付で、国立情報学研究所(NII)にて特任助教に着任しました。 越前研究室にて引き続き、信頼できるAIの研究に取り組んで参ります。よろしくお願いいたします。
I am pleased to share that I have joined the National Institute of Informatics (NII) as a Project Assistant Professor.
One of our co-authored papers has been accepted at CVPR 2026👏 Congrats 🎉
Turingインターン時に関わった共著論文が1本、CVPR 2026に採択されました。修士にして超優秀な@shojiro_yamabe さんの成果です🎉
#CVPR2026
My internship paper at Turing Inc. has been accepted to #CVPR2026 !🎉
"Text-Printed Image: Bridging the Image-Text Modality Gap for Text-centric Training of Large Vision-Language Models"
Big thanks to Turing, @tsubasashi, @futakw, and @onely7_deep for the support!
🎉 Thrilled to share that our three papers have been accepted to ICLR 2026! 🎉
⭐️ 2 first-author papers from Sakuma Lab
⭐️ 1 co-authored paper from Turing Inc.
Grateful to Prof. Sakuma and my labmates, and to @tsubasashi at Turing, for his mentorship and support!
🎉 WACV 2026 Accepted! 🎉
From my internship at CyberAgent AI Lab (@cyberagent_ai)
📢 We introduce Multimodal Adversarial Training (MAT), the first defense against multimodal (image+text) attacks, improving robustness across VL tasks.
📄 https://t.co/WX5HJrJalf
メンターを務めた共著論文 “Uncolorable Examples” が、APSIPA 2025にてBest Paper Finalist (< Top 2%) に選出されました🎉
Our paper “Uncolorable Examples” has been selected as a Best Paper Finalist (< Top 2%) at APSIPA 2025.
Paper:
https://t.co/lcr5cwjHSP