Summary
Keisuke Sakaguchi is a professor and NLP researcher with 14 years of experience bridging machine learning, psycholinguistics, and applied language technologies. He focuses on robust NLP for noisy text, commonsense knowledge, visual diagram understanding, language acquisition applications, and reliable evaluation—work that earned outstanding paper awards at ACL 2017 and AAAI 2020 and coverage in Forbes and MIT Technology Review. Trained at Johns Hopkins (PhD) with internships across industry labs (AI2, Microsoft Research Asia, IBM, ETS), he combines rigorous cognitive perspectives with practical system-building toward embodied AI that mirrors human efficiency. Based in Sendai, he currently holds faculty roles at Tohoku University, advises startups like Kotoba, and collaborates with RIKEN, reflecting a blend of academic leadership and industry impact. An underappreciated thread in his portfolio is a sustained interest in decoding algorithms and data-efficient collection methods that improve real-world robustness beyond benchmark gains.
14 years of coding experience
7 years of employment as a software developer
Master of Engineering (MEng), Computational Linguistics, Master of Engineering (MEng), Computational Linguistics at Nara Institute of Science and Technology
Master of Arts (MA), Psycholinguistics and Neurolinguistics, Master of Arts (MA), Psycholinguistics and Neurolinguistics at University of Essex
Johns Hopkins University
Bachelor of Arts (BA), Philosophy, Bachelor of Arts (BA), Philosophy at Waseda University
Japanese, English, Chinese