2025:

• All 3 of our submissions have been accepted at EMNLP 2025: 2 Main Conference papers and 1 Findings paper! See you in Suzhou!

• Our paper, “The World According to LLMs: How Geographic Origin Influences LLMs’ Entity Deduction Capabilities” has been accepted at COLM 2025. [ArXiv]

• Honoured to have recieved the COLING 2025 Best Dataset Paper Award, for our paper, NYT-Connections: A Deceptively Simple Text Classification Task that Stumps System-1 Thinkers.

• Our paper, Fine-Tuned LLMs are “Time Capsules” for Tracking Societal Bias Through Books has been accepted for publication at the main conference at NAACL 2025, April 29th to May 4th.

2024:

• Our 2 papers have been accepted for publication at the main conference at The 31st International Conference on Computational Linguistics (COLING 2025, January 19th to 24th), both nominated as Oral Presentations (<7% submitted papers)

• Honoured to have recieved the EMNLP 2024 Social Impact Paper Award.

• Our paper, STOP! Benchmarking Large Language Models with Sensitivity Testing on Offensive Progressions has been selected for Oral Presentation at EMNLP 2024.

• A pleasure to be invited to give the convocation address to the Brock University Graduating Class of 2024!

• Our 2 papers have been accepted for publication at the main conference at The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024, November 12-16). Almost all co-authors were undergrads. Way to go! Stay tuned for paper links.

• I’ve been invited to give a talk at the FirstOntario Performing Arts Centre, titled In Search of Digital Truth: How Large Language Models are Shaping Our Reality (And How We are Shaping Theirs!), Friday, November 1, at 7 PM. [Details + Get your tickets!]

• Honored to receive the 2024 Faculty of Mathematics and Science Excellence in Teaching Award (June).

Join Our Research Group at Emory

I am recruiting fully funded PhD students for Fall 2026 (applications due December 2025), as well as MS students interested in thesis research. I’m building a new interdisciplinary research group at the intersection of Natural Language Processing, AI Fairness, and Human-Centered AI.

Research Focus

We develop methods to make AI systems more fair, robust, and reliable while understanding their limitations and societal impacts. Current research directions include:

• Reasoning & Evaluation: Creating challenging benchmarks that push AI reasoning capabilities and reveal failure modes

• AI Safety: Developing methods to identify and mitigate harmful behaviors in language models, with focus on culturally-aware and context-dependent definitions of harm

• Interpretability: Opening the black box of AI systems to understand their decision-making and build trust

• Applied AI for Good: Leveraging AI for inclusive education, diverse storytelling, and equitable technology