Midori [緑] means “Green”. Our research philosophy is to develop AI that advances present human, economic, and societal needs without exhausting the computational, environmental, and institutional resources required by future generations. In this sense, sustainability guides both what we study and how we study it. 

We seek AI systems that are collaborative rather than centrally monopolized, responsible rather than vulnerable to misuse, minimalistic rather than wastefully over-engineered, and practically useful beyond standard benchmarks - especially for sustainability applications. These principles define Midori Lab’s research agenda around Minimalistic, Decentralized, and Responsible Intelligence.

RQ1) Minimalistic Intelligence: How can we do more with less? 

RQ2) Decentralized AI Challenge: How can we tackle the complexities associated with collaborative learning environments? 

RQ3) Responsible AI Challenge: How do we defend against malicious actors in an age where AI is for all to use, and easily weaponized? 

RQ4) Intelligent Applications: Beyond standard benchmarks, can we solve domain-specific problems with green intelligence? 

 🤓We particularly value works led by our own team members, or works originated at our lab by visiting students. 😆

For the below selected publications, *denotes our team members (my NTU students/staff) 

#denotes our visiting students (work completed during visiting attachment to our team at NTU)