PhD Student · Computer Science

Christabel Acquaye

University of Maryland, College Park
Advised by Prof. Rachel Rudinger · CLIP Lab

I am a fourth-year PhD student researching the development, fine-tuning, and evaluation of large language models to improve natural language understanding, with emphasis on commonsense reasoning, cultural alignment, question item generation and difficulty estimation, and fairness. My work spans cross-cultural benchmarking, LLM-based student simulation, and building trustworthy NLP systems that work equitably across diverse populations.

College Park, MD
Christabel Acquaye

Research Interests

Personalization Commonsense Reasoning Cultural Alignment Educational NLP Fairness & Bias Question Answering Item Response Theory Multimodal AI Cross-Cultural Evaluation LLM Adaptation

Advised by Prof. Rachel Rudinger at University of Maryland, College Park · CLIP Lab

News

July 2026
Paper presented at ACL 2026: Take Out Your Calculators: Estimating the Real Difficulty of Question Items with LLM Student Simulations and IRT.
July 2026
Paper accepted to ICPR 2026: All-in-One Conditioning for Text-to-Image Synthesis with Hirunima Jayasekara et al.
2025
Oral presentation at RTC Conference 2025: From Transcripts to Insights.
Jun 2025
Started NLP Research Internship at Vail Systems, Inc. in Chicago, IL.

Publications

* denotes equal contribution

ACL 2026 · Published
Take Out Your Calculators: Estimating the Real Difficulty of Question Items with LLM Student Simulations and IRT
Christabel Acquaye, Yi Ting Huang, Marine Carpuat, Rachel Rudinger
Educational NLPIRTLLM SimulationMath Assessment
ICPR 2026 · Accepted
All-in-One Conditioning for Text-to-Image Synthesis
Hirunima Jayasekara, Chuong Huynh, Yixuan Ren, Christabel Acquaye, Abhinav Shrivastava
Text-to-ImageScene GraphsMultimodalDiffusion Models
NAACL 2025 · Published
My LLM Might Mimic AAE — But When Should It?
Sandra Camille Sandoval*, Christabel Acquaye*, Kwesi Adu Cobbina*, Mohammad Nayeem Teli, Hal Daumé III
Language FairnessAAEEthicsLLM
EMNLP 2024 · Published
Susu Box or Piggy Bank: Assessing Cultural Commonsense Knowledge between Ghana and the US
Christabel Acquaye, Haozhe An, Rachel Rudinger
Cultural NLPCommonsenseBenchmarkGhana
ACL 2024 · Published
Do Large Language Models Discriminate in Hiring Decisions on the Basis of Race, Ethnicity, and Gender?
Haozhe An, Christabel Acquaye, Colin Wang, Zongxia Li, Rachel Rudinger
FairnessHiring BiasLLM Evaluation
Harvard Dataverse 2022 · Dataset
KaraAgroAI Cocoa and Maize Dataset
Darlington Akogo, Christabel Acquaye, Emmanuel Amoako, Jerry Buaba, Issah Samori, et al.
Cocoa: 1,518 downloads · Maize: 2,670 downloads
Agriculture AIComputer VisionOpen Data

Teaching

University of Maryland

Teaching Assistant Aug 2022 — Present
CMSC 470
Introduction to Natural Language Processing
Spring 2026
Instructor: Prof. Rachel Rudinger
CMSC 828J
Common-Sense Reasoning and Natural Language Understanding
Fall 2025 · Graduate Seminar
Instructor: Prof. Rachel Rudinger
CMSC 216
Introduction to Computer Systems
Summer 2023, Summer 2024
CMSC 320
Introduction to Data Science
University of Maryland
CMSC 131 / 132
Object-Oriented Programming I and II
University of Maryland

Academic City University College

Faculty Assistant May 2021 — Aug 2022
Emerging Technologies, Matlab for Algebra & Python Programming
Accra, Ghana

Kwame Nkrumah University of Science and Technology

Teaching Assistant Sep 2019 — Jun 2020
Computer Architecture, Semiconductor Devices, Engineering Technology, Basic Electronics, Embedded Systems, Digital Signal Processing
Kumasi, Ghana

ELiTE Education

Volunteer Tutor Jul 2016 — Jun 2019
Machine Learning & Hardware Programming
New York, USA
Tutored college and high school students in machine learning and hardware programming.

Recent Presentations

  • "From Transcripts to Insights: LLM-Assisted Topic Discovery" — RTC Conference 2025 (Oral)
  • "Susu Box or Piggy Bank: Cultural Commonsense Knowledge" — EMNLP 2024, Miami, FL (Poster)

Contact

I am happy to discuss research collaborations or any questions about my work. The best way to reach me is by email.

University of Maryland, College Park · College Park, MD