LAB LOGA page on LabLog, a review site (not the lab's official page)
Sakuma Lab
Machine-translated · View original
The lab studies the security and privacy of AI.
AboutAbout the lab
Toward trustworthy machine learning and artificial intelligence.
The Sakuma Lab studies the trustworthiness of machine learning and artificial intelligence.
It works on AI security and privacy, explainable AI that shows the reasons behind AI decisions, and AI-driven discrimination and fabrication.
The research covers a wide range of AI models, including language models, reinforcement learning, and models that link images and language.
Together with universities outside the lab's own, it also works on the pathological diagnosis of malignant lymphoma and on attacks against autonomous driving systems and countermeasures to them.
Key terms
- Misalignment
- A state in which an AI's behavior deviates from what people intended
- Red teaming
- A way of testing that takes the attacker's side to find the weak points of a system
- Explainable AI
- Technology that makes it possible to show, in a form people can understand, why an AI gave a particular answer
ResearchResearch themes
The lab studies the trustworthiness of machine learning and artificial intelligence, especially AI security and privacy, explainable AI, and AI-driven discrimination and fabrication. The research covers various AI models, including language models, reinforcement learning, and models that link images and language.
*Illustrative imageRed Teaming of Large Language Models
The lab is building a foundation for red teaming to uncover misalignment in large language models. The lab leads this project.
*Illustrative imageRobust Federated Foundation Models
The lab leads research on building federated foundation models that are less prone to breaking down, by generating synthetic data.
*Illustrative imageHow Attacks on AI Work, and Defenses Against Them
Drawing on the causal structure behind how AI distinguishes things, the lab works to understand how attacks on AI succeed and to design defensive measures.
*Illustrative imageSafety Evaluation of AI-Driven Systems
The lab participates as a principal collaborator in building a foundation for evaluating the security of AI-driven cyber-physical systems and developing countermeasures.
MembersFaculty
The members of the Sakuma Lab consist of faculty, researchers, secretaries, affiliated faculty from other institutions, doctoral, master's, and undergraduate students, and internship and exchange students.As of 2026
- ProfessorJun SakumaPublications ↗
- Assistant ProfessorRui WenPublications ↗
- Project Assistant ProfessorDaiki NishiyamaPublications ↗
VoicesStudent voices
Ratings and reviews reflect the experiences of individual current or former students, not the lab's official views. Whether a lab has reviews, or how many, says nothing about the lab itself.
Students who take part in an interview may receive a thank-you gift worth ¥1,000, regardless of what they say.
Student ratings (0 students)
--
--
--
--
--
--
Student voices answer questions like these
- What's great about this lab?
- What did you learn after joining?
- What's the vibe, and what is the professor like?
- Who is this lab for?
- Where do graduates go?
Your experience can help younger students choose their lab. Your name won't be shown.
Write a lab reviewAnonymous · Your draft is saved if you stopRelatedLabs similar to Sakuma Lab
Kurihara LabSchool of Computing, Department of Computer ScienceSecurity and privacy technologies, anonymous communication
Okazaki LabSchool of Computing, Department of Computer ScienceNatural language processing, machine learning (language anal…
Takayasu LabSchool of Computing, Department of Computer ScienceBig data analysis and statistical physics
Haruhiko Kaneko LabSchool of Computing, Department of Computer ScienceError control codes and data compression technologies (highl…The information on this page is compiled by the LabLog team from the lab's public information and interviews. For corrections or removal requests, please contact LabLog, not the lab.
Report an errorAbout the images
The top and research theme images are illustrative and were not taken at the lab. They are cropped from Wikimedia Commons images. Cropped CC BY-SA images are provided under the same CC BY-SA license as the original (version shown on each line).
- Top image: “Old Town Hall Bank Museum, vault door ( )” Preservation Maryland / CC BY-SA 2.0 (cropped)
- Red Teaming of Large Language Models: “WTBBL magnifier 01” Joe Mabel / CC BY-SA 3.0 (cropped)
- Robust Federated Foundation Models: “Jigsaw puzzle in progress” Balise42 / CC BY-SA 4.0 (cropped)
- How Attacks on AI Work, and Defenses Against Them: “Domino Cascade” aussiegall / CC BY 2.0 (cropped)
- Safety Evaluation of AI-Driven Systems: “20140619 Misiryeong Tunnel Control Center 02” Misiryeong / CC BY-SA 4.0 (cropped)
- Related lab (栗原研究室): “EthernetSwitch” Raysonho / CC0 (cropped)
- Related lab (岡崎研究室): “Birthdaytagcloud” User:Killiondude / Public domain (cropped)
- Related lab (高安美佐子研究室): “Market centre in Tokyo stock exchange” ehnmark / CC BY 2.0 (cropped)
- Related lab (金子研究室): “Hard disk head crash” Alchemist-hp / CC BY-SA 3.0 (cropped)
About LabLog
LabLog is a lab review app for students only. Student members at each university interview students in labs and publish what they say. Student voices can only be read by students who sign up with a university email address. Operated by bestiee Inc.
Learn more about LabLog ↗Search engines and AI may reference the basic lab info and rating summaries on this page and the description of LabLog. Student voices are for student members only. Automated collection by programs (scraping, or input to or training of AI) is prohibited under Article 8 of the Terms of Service. If you need data for research or analysis, please contact us.