Amirhossein Razavi

Amirhossein Razavi

Second-year MASc Student at Toronto Metropolitan University

Welcome to Amirhossein Razavi’s page

About Me

Welcome! I am Amirhossein Razavi, a second-year Master of Applied Science (MASc) student at Toronto Metropolitan University, where I conduct research under the supervision of Dr. Ebrahim Bagheri in the LS3 Lab. My research focuses on advancing the frontiers of Natural Language Processing (NLP) and Information Retrieval (IR), with particular emphasis on understanding and improving the Large Language Models (LLMs).

Research Interests

  • Natural Language Processing: LLMs, prompt engineering, RAG, prompt performance prediction
  • Information Retrieval: Query performance prediction, recommendation systems, search effectiveness

Benchmarking Prompt Sensitivity in Large Language Models
A. Razavi, M. Soltangheis, N. Arabzadeh, S. Salamat, M. Zihayat, E. Bagheri
European Conference on Information Retrieval (ECIR 2025)
Paper | Google Scholar

This research introduced the novel task of Prompt Sensitivity Prediction and presented PromptSET, a comprehensive benchmark dataset for investigating how variations in prompt formulation impact LLM performance. The work addresses critical challenges in deploying LLMs reliably in real-world applications. You can read more about this research in here.