MINT-NLP at MBZUAI
Welcome to the MINT at the Natural Language Processing Department of MBZUAI!
At MINT—short for MBZUAI Interpretability Team—
we explore the inner workings of NLP systems such as large language models (LLMs).
News!
2024/12/02 We have 2 papers accepted at COLING2025🎉
2024/11/23 One PhD student and four MSc students have joined our team as a new member!
Our Missions
Interpreting LLMs: Exploring the Mechanisms Behind How LLMs Process Natural Language
LLM's Knowledge
How do LLMs acquire and store the knowledge?
LLM's Inner Workings
How do LLMs solve tasks by utilizing the knowledge?
Human vs. LLM
How does NLP of human and LLMs differ?
Our Collaboration
We collaborate globally with leading institutions and other NLP teams at MBZUAI.
Notably, we partner with RIKEN and Tohoku NLP on projects that bridge language models, AI interpretability, and cognitive modeling.