Hello, I'm Shuyue.
Currently, my focus is on the intersection of large language models and multi-agent systems. Specifically, I am interested in: (i) improving the performance of LLMs by applying theories, principles, and methodologies from multi-agent systems, and (ii) using LLMs to address the open challenges within multi-agent systems.
What's new:
September 2025
Our paper "Beyond gpt-5: Making llms cheaper and better via performance-efficiency optimized routing" was accepted to DAI 2025!
September 2025
Our papers "Rema: Learning to meta-think for llms with multi-agent reinforcement learning" and "Scaling physical reasoning with the physics dataset" were accepted to NuerIPS 2025!
June 2025
Our paper "A formal model for multiagent Q-learning on graphs" was accepted to Science China Information Science 2025!
May 2025
Our paper "Regret minimization in population network games: vanishing heterogeneity and convergence to equilibria" was accepted to IEEE Transactions on Neural Networks and Learning Systems 2025!
April 2025
Our paper "Understanding When and Why Graph Attention Mechanisms Work via Node Classification" was accepted to ICML 2025!
January 2025
Our paper "Payoff Control in Multichannel Games: Influencing Opponent Learning Evolution" was accepted to IEEE Transactions on Cybernetics!