I am a PhD student at the University of Illinois Urbana-Champaign, advised by Dilek Hakkani-Tur.
My research centers on natural language processing and large language models, where I'm particularly interested in conversational AI, evaluation, and synthetic data generation. Recent works include user simulation, retrieval-augmented generation, tool-calling, instruction-tuning, persuasion, and discourse parsing. I am currently a research scientist intern at IBM Research.
Before starting my PhD, I earned my BSc from the University of British Columbia, where I was fortunate enough to work with professors Vered Shwartz and Giuseppe Carenini. Additionally, I completed internships at iClinic Systems, Amazon (x3), and Clir Renewables.
Some Recent News
| 05/2025 | Starting my Research Scientist Internship at IBM Research in San Jose. |
|---|---|
| 04/2024 | I accepted my PhD offer from UIUC CS! Very excited to start in Fall 2024. |
| 01/2024 | 🎉 My submission has been accepted at CODI 2024, the workshop on Computational Approaches to Discourse, part of EACL 2024. |
| 01/2024 | I was awarded the Mitacs Accelerate Fellowship for Master's and PhD Students as an undergraduate student. Today is my first day at iClinic Systems Inc., I'm excited to start working on information extraction in a clinical setting! |
| 10/2023 | 🎉 My paper has been accepted at EMNLP 2023! Looking forward to attending my first research conference and presenting in Singapore! |
Publications
-
Goal Alignment in LLM-Based User Simulators for Conversational AI [Paper]
Shuhaib Mehri, Xiaocheng Yang, Takyoung Kim, Gokhan Tur, Shikib Mehri, Dilek Hakkani-Tür.
Preprint. -
Must Read: A Systematic Survey of Computational Persuasion [Paper]
Nimet Beyza Bozdag, Shuhaib Mehri, Xiaocheng Yang, Hyeonjeong Ha, Zirui Cheng, Esin Durmus, Jiaxuan You, Heng Ji, Gokhan Tur, Dilek Hakkani-Tür.
Preprint. -
PIPA: A Unified Evaluation Protocol for Diagnosing Interactive Planning Agents [Paper]
Takyoung Kim*, Janvijay Singh*, Shuhaib Mehri*, Emre Can Acikgoz, Sagnik Mukherjee, Nimet Beyza Bozdag, Sumuk Shashidhar, Gokhan Tur, Dilek Hakkani-Tür.
Preprint. -
Persuade Me if You Can: A Framework for Evaluating Persuasion Effectiveness and Susceptibility Among Large Language Models [Paper]
Nimet Beyza Bozdag, Shuhaib Mehri, Gokhan Tur, Dilek Hakkani-Tür.
Preprint. -
Beyond Sample-Level Feedback: Using Reference-Level Feedback to Guide Data Synthesis [Paper]
Shuhaib Mehri, Xiusi Chen, Heng Ji, Dilek Hakkani-Tür.
Preprint. -
Automatic Evaluation of Generative Models with Instruction Tuning [Paper]
Shuhaib Mehri, Vered Shwartz
EMNLP 2023 GEM Workshop.