Eric Onyame
Eric Onyame
View CV

About Me

I am a PhD candidate in Data Science at the University of Virginia School of Data Science, where I am fortunate to work with Dr. Chirag Agarwal in the AIKYAM Lab. My research focuses on Trustworthy AI, with particular emphasis on AI Safety, interpretability, and Multilingual AI. Broadly, I study methods for understanding and monitoring large language models to support scalable oversight, especially when model behavior changes under distribution shift.

Before starting my PhD, I earned an MS in Mathematics from the University of Tennessee at Chattanooga, where I worked with Dr. Lakmali Weerasena on facility location problems in mathematical optimization. I received my bachelor's degree in Mathematics with Economics (First Class Honors) from the University of Cape Coast in Ghana.

Current Research

I am currently studying interpretability for AI safety, with a focus on chain-of-thought (CoT) monitoring and how white-box interpretability methods can make it more reliable. My work explores how we can monitor model reasoning to detect and mitigate deceptive, unfaithful, and other misaligned behaviors in large language models, with the broader goal of supporting scalable oversight.

Recent works include:

  • The Fragility of Chain-of-Thought Monitoring Across Typologically Diverse Languages (arXiv 2026).
  • Curriculum-informed reinforcement learning for multilingual medical reasoning (ACL 2026, Oral).

News

Recent updates and milestones.

Apr 2026
One paper on multilingual medical reasoning in LLMs accepted to ACL 2026 (Main Conference, Oral). See you in San Diego! 🎉
Jun 2025
Passed the qualifying exam. 🎉
Mar 2025
Gave a talk on applications of large language models at the UVA School of Data Science.
Aug 2023
Started my PhD at the University of Virginia School of Data Science. 🎉
Mar 2023
Received a fully funded fellowship and the Provost Scholarship to pursue my PhD at the University of Virginia, and was also selected as a Quantitative Foundation Fellow.
Mar 2023
Received the 2023 Outstanding Master's Student in Mathematics award at UTC. 🎉
Feb 2023
Defended master's thesis: Covering Problem with Minimum-Radius Enclosing Circle.
Apr 2022
Inducted into Pi Mu Epsilon Mathematics Honor Society.

Research

I spend most of my time thinking about how to make large language models more reliable, interpretable and aligned with human values. I am broadly interested in Trustworthy AI, with a focus on developing methods for monitoring model behavior, detecting failures and supporting scalable oversight under distribution shift. I am motivated by a central question: how can we build AI systems whose reasoning and outputs remain faithful, safe, and trustworthy across changing tasks, languages, and contexts?

Recently, I have been especially interested in three directions:

Chain-of-Thought Monitoring and Scalable Oversight

How can we monitor the reasoning traces of large language models to better understand when they are faithful, when they fail, and when they may conceal unsafe behavior? I am interested in chain-of-thought monitoring as a tool for scalable oversight, especially for detecting and mitigating unfaithful, deceptive, and scheming behavior in advanced AI systems.

AI Safety and Interpretability

How can we use interpretability methods to understand, evaluate, and improve the safety of AI systems? I study approaches for detecting unsafe or misleading model behavior, with a particular interest in how white-box interpretability methods can make monitoring and oversight more reliable.

Multilingual AI under Distribution Shift

How can we build and evaluate models that remain reliable when language, culture, or context changes? I study how model behavior shifts across linguistic settings, with the goal of making AI systems more robust, faithful, and trustworthy for diverse users.

Publications

Selected publications and ongoing work.

White-Box Approaches for Chain-of-Thought Monitorability
Eric Onyame, Chirag Agarwal.
In progress
Counterfactual LLM Verifiers for Math and Logic Reasoning Tasks
Elita Lobo, Eric Onyame, Yair Zick, Chirag Agarwal.
In progress

Education

Academic journey.

University of Virginia
PhD in Data Science · Aug 2023 – Present
Advisor: Dr. Chirag Agarwal
The University of Tennessee at Chattanooga
MS in Mathematics · Aug 2021 – May 2023
Advisor: Dr. Lakmali Weerasena
University of Cape Coast, Ghana
BSc in Mathematics with Economics · Aug 2016 – Jul 2020
First Class Honors

Teaching

Teaching Assistant for core Data Science and AI courses.

Spring 2026
DS 6050: Deep Learning
Led office hours, assisted with assignments, and guided students on optimization, regularization, and neural network architectures.
Fall 2025
DS 7800: Research Methods in Data Science
Supported student research development, methodology design, and critical evaluation of data-driven studies.
Spring 2025
DS 6051: Decoding Large Language Models
Assisted instruction on modern language models, including projects, readings, and practical evaluation and prompting workflows.
Fall 2024
DS 6600: Data Engineering I
Facilitated hands-on sessions on large-scale data systems and data visualization pipelines.
Fall 2024
DS 1001: Foundations of Data Science
Supported student understanding of core data science concepts through discussion, problem-solving, and feedback.
2021 – 2022
Elementary Statistical Analysis; College Algebra (UTC)
Delivered lectures, graded assessments, and provided structured academic support to undergraduate students.

Skills

Technical proficiencies.

Languages
Python R Julia SQL
Tools & Libraries
PyTorch PySpark Docker Git NumPy Pandas Matplotlib scikit-learn LaTeX Cloud Computing