Ehsan Latif

Ehsan Latif

AI Research Scientist

I build AI systems that reason, collaborate, and make meaningful impact. As an AI Research Scientist, I explore how LLMs, multi-agent intelligence, and trustworthy AI can move from strong benchmarks to real-world outcomes in education and robotics.

My current work explores reliable agentic workflows, AI-augmented assessment, and human-centered evaluation of next-generation models. Across NSF/IES-funded projects, I train and deploy transformer models (BERT, GPT, LLaMA), design coordination and localization algorithms, and study how intelligent systems can support learning at scale.

My research has been recognized at AIED, NeurIPS, AAAI, IROS, and ICRA. If you are interested in collaborative research on trustworthy LLMs, AI for education, or multi-agent systems, I would love to connect and build together.

Selected Publications

A systematic review of intelligent and robot tutoring systems: evolution, pedagogical design, and AI-driven classification New

E. Latif, V. Liu, & X. Zhai
Smart Learning Environments
December 2026

A comprehensive review of intelligent and robotic tutoring systems, covering historical evolution, pedagogical design choices, and AI-centered taxonomy.

Comparative evaluation of OpenAI O1 and human performance in higher order cognition New

E. Latif, Y. Zhou, S. Guo, Y. Gao, L. Shi, M. Nyaaba, A. Bewersdorff, X. Yang, ...
Scientific Reports
December 2025

Benchmarks OpenAI O1 against human performance on higher-order cognition tasks and analyzes implications for AI-supported assessment.

Latest News

December 1, 2026

New Smart Learning Environments publication

"A systematic review of intelligent and robot tutoring systems: evolution, pedagogical design, and AI-driven classification" authored by Ehsan Latif, V. Liu, & X. Zhai — published in Smart Learning Environments.

Read more
January 28, 2026

International Journal of Science Education publication

"Transforming science teacher education in the era of generative artificial intelligence: global perspectives" co-authored by Ehsan Latif and collaborators — published in International Journal of Science Education.

Read more
December 26, 2025

Scientific Reports paper on OpenAI O1 evaluation

"Comparative evaluation of OpenAI O1 and human performance in higher order cognition" authored by Ehsan Latif and collaborators — published in Scientific Reports.

Read more