Lucia Cipolina Kun

I am a Research Scientist at META’s FAIR lab, where I work on large language model (LLM) agents for scientific discovery. My research focuses on developing novel techniques for the self-improvement of LLM agents and their application to coding and reasoning. I develop frameworks combining reinforcement learning, multi-agent systems, and game theory to study emergent behaviors in post-trained LLMs, with a focus on designing systematic evaluation protocols that assess their strategic reasoning, planning, and behavioral stability.

My professional background includes roles as a Student Researcher at Google DeepMind, where I developed novel techniques for multi-agent decision processes, and as a Research Intern at Juelich Supercomputing Center, focusing on high-performance computing, large-scale generative models and LLM evaluation. I have also interned at JPMorgan AI Research optimizing reinforcement learning algorithms for trading, and at Microsoft AI, working on model-based offline reinforcement learning for Bing’s search and revenue management. Earlier in my career, I worked as a front office quant at Morgan Stanley, Bank of America and HSBC in NY, implementing machine learning and diffusion models in C++.

My education includes an MSc in Mathematics from the Courant Institute at the New York University, where I specialized in stochastic differential equations and a PhD in Electrical Engineering at the University of Bristol.

I am also involved in a side project exploring AI-based art restoration, where I investigate how machine learning can help preserve and restore cultural heritage.