Diego Antognini

I am Diego Antognini, a research scientist at IBM Research AI. I have over 6 years of research experience in natural language processing (NLP), machine learning (ML), and single- and multi-objective recommendation systems. I am currently working on machine learning and natural language processing. My research is twofold. First, I am developing new methods to align and personalize large language models (LLMs) according to user preferences. Second, I am interested in efficient machine learning for NLP. I build models with a model size on the order of one megabyte and a latency of one millisecond, or models that achieve significantly faster training, fine-tuning, or inference times than their larger counterparts, resulting in suitable models for resource-constrained embedded systems and data centers. I am also interested in different pre-training strategies to achieve high-quality or lower-cost pre-training. I have experience in interpretable models that generate personalized and actionable textual explanations (see below for more information).

Additionally, I am a lecturer and module head at the Lucerne University of Applied Sciences (HSLU) where I teach deep learning for natural language processing classes.

I hold a Ph.D. degree in Computer Science from the Swiss Federal Institute of Technology in Lausanne (EPFL), where I conducted research in the Artificial Intelligence Laboratory (LIA) under the supervision of Professor Boi Faltings. My doctoral thesis is titled "Textual Explanations and Critiques in Recommendation Systems". During my PhD, I developed models to infer high-quality explanations from text documents in a scalable and data-driven manner through selective rationalization. Additionally, I designed new models to make explanations actionable (referred to as critiquing) and explored two important applications in natural language processing and conversational recommendation systems. I also worked on multi-objective optimization in recommendation systems and multi-document summarization.