Rose Niousha

Rose Niousha

PhD Student · UC Berkeley EECS

I am a fourth-year PhD student in Electrical Engineering and Computer Sciences at UC Berkeley, advised by Prof. Narges Norouzi.

I am driven by the question: How do humans interact with and learn alongside AI systems? My research focuses on modeling messy, real-world human behavior and developing scalable methods for evaluating AI assistants. I currently explore these ideas in intelligent tutoring, where I build LLM-based user simulators that capture not only what learners do, but how they reason.

I am also the president of Computer Science Graduate Entrepreneurs (CSGE) at Berkeley, where we support CS PhD students interested in turning their research into startups.

Prior to my PhD, I completed my Bachelor of Engineering in Computer Science and Communications Engineering at Waseda University in Tokyo, Japan, where I also grew up.

Outside of academia, I enjoy cooking and weightlifting :)

Recent News

Selected Publications

EMNLP · 2026
ParaStudent: Closing the Sim2Real Gap in User Simulators for AI Tutor Evaluation
Rose Niousha, Mihran Miroyan, Abigail O'Neill, Joseph E. Gonzalez, Gireeja Ranade, John DeNero, Narges Norouzi
We close the sim2real gap in LLM student simulators, making them reliable enough to evaluate AI tutors
Paper Code
COLM · 2026
INSIDE the Student's Mind: Jointly Modeling Latent Reasoning and Action in LLM Student Simulators
Rose Niousha*, Minwoo Kang, Narges Norouzi
We model students' latent reasoning jointly with their actions, so simulators think like learners rather than just imitating them
Paper Code
AIED · 2026
The Missing Evaluation Axis: What 10,000 Student Submissions Reveal About AI Tutor Effectiveness
Rose Niousha, Samantha Boatright Smith, Bita Akram, Peter Brusilovsky, Arto Hellas, Juho Leinonen, John DeNero, Narges Norouzi
AI tutors should be evaluated beyond pedagogical soundness
Paper
AAAI · 2025
Comparing Artificial Intelligence Curricula in Canadian and US Universities
Rose Niousha, Lexie Jingruo Guo, Rick Kaifeng Li, Narges Norouzi, Lisa Zhang
Canadian AI courses sit behind longer prerequisite chains than US ones, delaying student access to AI
Paper
SIGCSE · 2024
Use of Large Language Models for Extracting Knowledge Components in CS1 Programming Exercises
Rose Niousha, Muntasir Hoq, Bita Akram, Narges Norouzi
LLMs recover CS1 knowledge components that closely match an expert-generated list
Paper
Education Sciences · 2023
Investigating the Effect of Binary Gender Preferences on Computational Thinking Skills
Rose Niousha, Daisuke Saito, Hironori Washizaki, Yoshiaki Fukazawa
Scratch project type biases computational thinking scores, confounding measured gender differences
Paper