Grgur Kovač

Hello, I am a PhD student in the Flowers Team (INRIA, Bordeaux). My research concerns the socio-cultural aspects of AI and how psychology and human sciences can be leveraged to better characterize, evaluate, and build LLMs and RL agents.

Projects

You can see our projects below:

• Recursive Training Loops in LLMs - How training data properties modulate distribution shift in generated data?
• Stick to your Role! Leaderboard - a leaderboard based on and extending the paper
• Stick to your Role! Stability of Personal Values Expressed in Large Language Models (paper) - a study of Personal Value stability in LLMs
• Large Language Models as superpositions of cultural perspectives - a positioning regarding the influence of context on the expression of culture in LLMs
• The SocialAI school - a perspective based on developmental psychology, and a tool facilitating research into socio-cognitive abilities of AI agents
• GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning - study of intrinsically motivated goal-based exploration with RL

Other projects:

• When LLMs Play the Telephone Game - Cumulative Changes and Attractors in Iterated Cultural Transmissions
• Autotelic LLM-based exploration for goal-conditioned RL - open-ended exploration with LLMs and RL

See my papers here

Contact

email: grgur.kovac@inria.fr