Studying LLMs through a computational social science lens
Drawing on classical social influence theory and my HCI background, I study how LLMs behave under peer pressure and social influence in multi-agent settings (preprint, preprint).
Jiaming Qu
Applied Scientist, Amazon / Ph.D. in Information Science, UNC Chapel Hill
Hello! My name is Jiaming Qu (瞿佳明) and I am now an applied scientist at Amazon. I got my Ph.D. degree in Information Science at School of Information and Library Science (SILS), UNC Chapel Hill, where I was advised by Prof. Yue Wang and Prof. Jaime Arguello. I also got my Master's degree in Information Science at SILS.
My research lies at the intersection of Explainable Artificial Intelligence (XAI), Human-Computer Interaction (HCI), and Information Retrieval (IR). I develop explainable AI systems with a focus on transparency, responsibility, and usability, and I conduct empirical studies from a human-centered perspective to understand user behavior and guide system design.
Publications
Preprints & Under Review
Most LLM Conformity Needs No Speaker: Measuring the Speaker-Free Floor in Peer-Pressure Benchmarks.
arXiv preprint.
Easier to Mislead Than to Correct: Harmful and Beneficial Revision in LLM Conformity.
arXiv preprint.
AGWM: Affordance-Grounded World Models for Environments with Compositional Prerequisites.
arXiv preprint.
Touching Space: Accessible Map Exploration Through Conversational Audio-Haptic Interaction.
arXiv preprint.
Published & Accepted Papers
✦ Served as corresponding author to mentor junior researchers.
Possible or Definite? A Benchmark for Evaluating Diagnostic Uncertainty Preservation in Clinical Text.
Machine Learning for Healthcare Conference (MLHC 2026).
Why is “Chicago” Predictive of Deceptive Reviews? Using LLMs to Discover Language Phenomena from Lexical Cues.
ACL 2026 Workshop on Trustworthy Natural Language Processing (TrustNLP 2026).
From Frontier to Frugal: Evaluating Self-Evolution Frameworks with Small Language Models.
ACL 2026 Workshop on Sustainable and Efficient Language, Vision, and Action Models (SELVA 2026).
A Multi-Agent Framework for Democratizing XR Content Creation in K-12 Classrooms.
HCI International 2026.
Understanding the Effects of Explaining Predictive but Unintuitive Features in Human-XAI Interaction.
Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency.
Why is “Problems” Predictive of Positive Sentiment? A Case Study of Explaining Unintuitive Features in Sentiment Classification.
Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency.
Understanding the Cognitive Influences of Interpretability Features on How Users Scrutinize Machine-Predicted Categories.
Proceedings of the 2023 Conference on Human Information Interaction and Retrieval.
A Study of Explainability Features to Scrutinize Faceted Filtering Results.
Proceedings of the 30th ACM International Conference on Information & Knowledge Management.
A Deep Analysis of an Explainable Retrieval Model for Precision Medicine Literature Search.
Advances in Information Retrieval: 43rd European Conference on IR Research.
Towards Explainable Retrieval Models for Precision Medicine Literature Search.
Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval.
Dissertation & Technical Reports
Explaining Unintuitive Feature Importance Explanations.
Ph.D. Dissertation, University of North Carolina at Chapel Hill, 2025.
A Medical Literature Search System for Identifying Effective Treatments in Precision Medicine.
Master's Thesis, University of North Carolina at Chapel Hill, 2019.
Papers I Did During Undergrad and Master
Should I Invest it? Predicting Future Success of Yelp Restaurants.
Proceedings of the Practice and Experience on Advanced Research Computing.
Urban Mining Potentials of University: In-use and Hibernating Stocks of Personal Electronics and Students' Disposal Behaviors.
Resources, Conservation and Recycling.
RAFFIA: Short-term Forest Fire Danger Rating Prediction via Multiclass Logistic Regression.
Sustainability.
Experience
PC Member / Reviewer
IR / AI / NLP venues
CIKM2026
SIGIR2024, 2025, 2026
WWW2023, 2026
WSDM2026
ICTIR2023
ARR2025 Feb, 2026 Jan, 2026 May
HCI venues
CHI2024 (outstanding review recognition), 2026
IUI2026
DIS2026
CUI2026
UbiComp2026
MobileHCI2026
Volunteer: WWW'23, CHIIR'23.
Work experience
May 2025 – present
Applied Scientist
Seattle
Amazon
I work on the Customer Service Data Intelligence team. I develop and evaluate LLM-based approaches for understanding why customers contact Amazon, in order to improve customer service quality.
Aug 2019 – May 2025
Research Assistant
Chapel Hill
School of Information and Library Science, UNC Chapel Hill
I worked with Prof. Yue Wang, Prof. Jaime Arguello and Prof. Rob Capra across multiple research projects on information retrieval, human-computer interaction and explainable AI.
May 2023 – Aug 2023
Applied Scientist Intern
Sunnyvale
Amazon
I worked on the Local Information team in Alexa AI. I developed a LambdaMART learning-to-rank model to improve Alexa's local business search in the Japanese market. The model was launched in production, leading to a 4% increase in customer satisfaction within one week and impacting over 1 million queries annually.
May 2022 – Aug 2022
Applied Scientist Intern
Sunnyvale
Amazon
I worked on the Local Information team in Alexa AI. I developed a BERT-based classification model to predict Alexa search query intentions, boosting Precision@1 by 14% and Recall@1 by 51%. The model was deployed within the team's pipeline, providing predictions as features for downstream tasks.
Education
Aug 2019 – May 2025
Ph.D., Information Science
University of North Carolina at Chapel Hill
Dissertation: Explaining Unintuitive Feature Importance Explanations.
Aug 2017 – May 2019
M.S., Information Science
University of North Carolina at Chapel Hill
Thesis: A Medical Literature Search System for Identifying Effective Treatments in Precision Medicine.
Award: Dean's Achievement Award for the best Master's Thesis (2 out of 95; first Chinese student recipient in the award's history).
Sep 2013 – May 2017
B.S., Management Information Systems
Nanjing Forestry University
Award: Outstanding Bachelor's Thesis (top 3%).
Contact
Email: qjiaming [at] amazon [dot] com
I am always interested in doing research in explainable AI, human-AI interaction, information retrieval, and LLMs. I am open to collaboration opportunities, and I especially enjoy working with interdisciplinary researchers and mentoring junior researchers.
Disclaimer: all work presented here is self-funded and conducted outside employer hours; opinions expressed are solely my own and do not represent Amazon.