Tonglin Yan

I am a third-year Ph.D. student at the CIAMS, Université Paris-Saclay, supervised by David Rudrauf (CIAMS, Université Paris-Saclay), Alain Finkel (LMF, ENS Paris-Sacaly), and Grégoire Sergeant-Perthuis (LCQB, Sorbonne Université).

Research Interests

My research investigates human-like intelligence in virtual agents through the integration of computational consciousness models, specifically the Projective Consciousness Model (PCM), with LLMs. The core of my work focuses on developing methods that fuse cognitive data—including facial expressions, gaze patterns, bodily movements, and linguistic content—within a unified world model to generate coherent and contextually appropriate verbal and nonverbal behaviors in immersive virtual reality environments. I am also interested in computational consciousness models as a scientific framework for understanding human consciousness, with a particular emphasis on linking computational theory and implementable simulations. A broader research interest extends to the evaluation of LLMs within social cognitive scenarios, identifying specific limitations in their social reasoning capabilities, such as theory of mind and context-sensitive decision-making, thereby advancing the development of more sophisticated models for socially situated virtual agents.