Hi! 👋 I’m Davide Gabrielli, a PhD student in Computer Science at Sapienza University of Rome. My main research focus is the adversarial robustness of Large Language Models (LLMs), investigating how to better understand and improve their behavior in the presence of non-sensical or malicious inputs.

In parallel, I’m also developing interpretable AI systems for anomaly detection in physiological time series data, with applications in healthcare. This work involves using LLMs to generate natural language explanations for complex model behavior, helping bridge the gap between AI and human decision-making.

More broadly, I’ve worked on projects spanning deep learning, generative models, and audio AI, including interpretable sound synthesis using Differentiable Digital Signal Processing (DDSP). My overarching goal is to contribute to building trustworthy, robust, and understandable AI systems.