Portrait of Hugh Mee Wong

Hugh Mee Wong

Amsterdam, The Netherlands

Hi, I am Hugh Mee. I am a PhD candidate in the Natural Language Processing group at Utrecht University, advised by Prof. Albert Gatt and Dr. Rick Nouwen. My work is funded by the Dutch Research Council (NWO) through an AiNed Fellowship grant awarded to Prof. Massimo Poesio.

Recently, my main research has focused on using mechanistic interpretability methods to understand language model behaviour. Ultimately, I would like to work towards more actionable forms of interpretability, where our understanding of model internals can help us identify problematic behaviour, evaluate models in a more meaningful way, and develop methods that make language models more safe and reliable. In my current research projects, I study model internals to monitor signals that might not be apparent from model behaviour alone.

I also occasionally work on multimodality, where I am interested in the different roles that individual modalities play in learned representations and how information is integrated across modalities. Here, I also draw inspiration from psycholinguistics, as I think studying multimodal models may also help us understand how humans process and combine different kinds of information (and maybe vice versa).

I obtained my MSc degree in Logic in 2023 and my BSc degree in Artificial Intelligence in 2020, both from the University of Amsterdam.

Email: h♗m♗wong♔uu♗nl (where the bishop is a dot and the king is an at symbol)

LinkedIn Google Scholar ACL Anthology

Publications

Investigating Multimodal Informativity under Different Partner Visibility Conditions in Video-Mediated Dialogue
Esam Ghaleb*, Hugh Mee Wong*, Kristina Kobrock
Preprint
From Vision to Language: Investigating Causal Information Flow in Multimodal Decision-Making
Davide Testa, Hugh Mee Wong, Alessandro Lenci, Bernardo Magnini, Albert Gatt
Forthcoming, Findings of EMNLP 2026
When Models Decide and When They Bind: A Two-Stage Computation for Multiple-Choice Question Answering
Hugh Mee Wong, Rick Nouwen, Albert Gatt
Findings of the Association for Computational Linguistics (Findings of ACL 2026)
Disentangling the Roles of Representation and Selection in Data Pruning
Yupei Du, Yingjin Song, Hugh Mee Wong, Daniil Ignatev, Albert Gatt, Dong Nguyen
Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)
VAQUUM: Are Vague Quantifiers Grounded in Visual Data?
Hugh Mee Wong, Rick Nouwen, Albert Gatt
Findings of the Association for Computational Linguistics (Findings of ACL 2025)
DeMeVa at LeWiDi-2025: Modeling Perspectives with In-Context Learning and Label Distribution Learning
Daniil Ignatev, Nan Li, Hugh Mee Wong, Anh Dang, Shane Kaszefski Yaschuk
Proceedings of the The 4th Workshop on Perspectivist Approaches to NLP (NLPerspectives 2025)
Attention vs non-attention for a Shapley-based explanation method
Tom Kersten, Hugh Mee Wong, Jaap Jumelet, Dieuwke Hupkes
Proceedings of Deep Learning Inside Out: The 2nd Workshop on Knowledge Extraction and Integration for Deep Learning Architectures (DeeLIO 2021)

Service

Academic Service

Departmental Service

Teaching

Utrecht University

University of Amsterdam

Miscellaneous (Personal Interests)