Anna Vettoruzzo

Hello! I am a postdoctoral researcher in the OpenEuroLLM and ELLIOT project within the AMOR/e group at Eindhoven University of Technology, where I focus on advancing the adaptability, resource efficiency, and generalization capabilities of foundation models. Core methodologies include developing parameter-efficient fine-tuning (PEFT) strategies, continual learning frameworks, and designing novel benchmarks to evaluate agentic and interactive intelligence.

I completed my Ph.D. in computer science at Halmstad University, Sweden, as part of the Center for Applied and Intelligent Systems Research (CAISR). My doctoral research focused on enhancing the generalization abilities of machine learning models through meta-learning, exploring how models can learn to learn. My work aimed to develop algorithms that perform well across diverse tasks with limited data - an essential step toward advancing deep learning models to better mimic human learning abilities.

Beyond my core research areas, my interests include world models, continual learning, self-supervised learning, and in-context learning.