About
Hi,
I’m a Postdoctoral Research Fellow at the MilaNLP group at Bocconi University. My research lies in understanding language models facilitate information. For that, I investigate biases embedded in large language models (LLMs), how they manifest in downstream tasks, and how they are encoded during the training phases of LLMs. I’m also interested in understanding the role of LLMs in the information environment in the context of information seeking. In the past, I worked with developing methods to mine ideologies, and more generally, opinions from texts.
My research interests naturally lead me to reflect a lot about the impact of AI systems in society – in particular how generative language models should normatively behave. Reach out if you like discussing that!
Finally, I also enjoy thinking about the implementation of my research in real world use case applications for social good, such as news recommenders. Speaking of which, I am co-leading the project MULTIVIEW which investigates methods for diversifying news recommendations in terms of perspectives.
In my free time, I like to read, bike and free-dive. :)