Scholar
Mingqian Zheng
Google Scholar ID: YAiRrrUAAAAJ
Language Technologies Institute, Carnegie Mellon University
Citations & Impact
Citations
109
H-index
1
i10-index
1
Publications
4
Co-authors
9
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'Let Them Down Easy! Contextual Effects of LLM Guardrails on User Perceptions and Preferences' was mentioned in a Forbes article on AI welfare and accepted to EMNLP 2025; another paper 'Synthetic Socratic Debates: Examining Persona Effects on Moral Decision and Persuasion Dynamics' also accepted to EMNLP 2025; gave an invited talk at Pareto.ai about recent work on LLM refusals; first paper got accepted to the Findings of EMNLP 2024.
Research Experience
- Conducted NLP research with David Jurgens in the Blablablab and worked with David Flood as a member of the HPACC team at the University of Michigan.
Education
- Ph.D. student in LTI at Carnegie Mellon University, co-advised by Carolyn Rosé and Maarten Sap; Master's in Survey and Data Science from the University of Michigan, advised by Yajuan Si; Bachelor's at NYU Shanghai with double majors in Mathematics and Data Science (Computer Science Concentration), worked with Hongyi Wen on Recommendation Systems.
Background
- Research interests include the dynamics of communication between humans and Large Language Models (LLMs), as well as interactions among multiple LLMs. The goal is to optimize these conversations for better human-AI collaboration. Research also examines LLM safety through refusal strategies, multi-turn interaction risks, and alignment between user expectations and model behavior. Additionally, studies multi-agent social simulations to uncover LLMs' behavioral patterns and social reasoning in complex social contexts.
