About Me

Dr. Giselle Zeno is an Artificial Intelligence Researcher at MIT Lincoln Lab. Her research interests lie in the fields of machine learning and data mining, particularly focusing on graphs, temporal processes, and generative models. Her primary work involves developing and analyzing algorithms for relational domains, including social, information, and communication networks, with applications to real-world tasks. More recently, her work has focused on evaluating large multimodal models, including a study of selection bias in multiple-choice probes of spatial understanding that received the Best Paper Award at the CVPR 2025 BEAM workshop, and on making models robust to natural corruptions. She has also served as organizing chair of the Synthetic & Adversarial ForEnsics (SAFE) workshop at WACV 2026 and CVPR 2026, and is an area chair for its next edition at AAAI 2027.