Miriam Doh

Visiting Scientist

miriamdoh.en.md photo

Miriam Doh is a doctoral researcher jointly affiliated with the ISIA Lab at Université de Mons (UMONS) and the IRIDIA Lab at Université Libre de Bruxelles (ULB). Her research focuses on the transparency and reliability of AI systems in facial analysis, with an emphasis on biases and ethical implications. Her work adopts an interdisciplinary approach, integrating principles of cognitive psychology with insights from legal studies, gender studies, and policy to examine the societal impacts of AI. During her three-month research visit to Ellis Alicante (November 2024–February 2025), she will study the Halo Effect’s cognitive bias of attractiveness in generative AI for human face generation.

Publications in association with ELLIS Alicante

2025

07/17
Vancouver, CA
Doh, M., Höltgen, B., Riccio, P., & Oliver, N. (2025). Position: The Categorization of Race in ML is a Flawed Premise. ICML'25: International Conference on Machine Learning.
Poster Spotlight
06/30
Eindhoven, NL
Doh, M., Gulati, A., Mancas, M., & Oliver, N. (2025). When Algorithms Play Favorites: Lookism in the Generation and Perception of Faces. European Workshop on Algorithmic Fairness.
06/23
Athens, GR
Doh, M., Canali, C., & Oliver, N. (2025). What TikTok Claims, what Bold Glamour Does: a Filter´s Paradox. ACM Conference on Fairness, Accountability and Transparency, FAccT 2025.
04/27
Yokohama, JP
Doh, M., Canali, C., & Oliver, N. (2025). Filters of Identity}: AR Beauty and the Algorithmic Politics of the Digital Body. CHI'25: International Conference on Human-computer Interaction. Workshop on Body Politics: Unpacking Tensions and Future Perspectives for Body-Centric Design Research in HCI.
02/25
Philadelphia, US
Doh, M., Gulati, A., & Oliver, N. (2025). Attractive by Design: How The Attractiveness Halo Effect Shapes AI Perception. Collaborative AI and modeling of Humans (CAIHu) - Bridge program at AAAI 2025.