research HOLA-Drone Hypergraphic open-ended learning for zero-shot multi-drone cooperative pursuit. Metacognitive Agent Test-time reasoning and adaptation for agents that learn to adjust like humans. Multi-Agent Cooperation Aligning individual and collective objectives in cooperative multi-agent systems. Cross-Utterance Speech Generation Coherent speech generation and editing with cross-utterance conditioned latent models. platform COLE Platform Cooperative open-ended learning for zero-shot coordination and human-AI teamwork. JiangJun Tackling non-transitivity in two-player zero-sum games through Xiangqi agents. science Embodied Science Agent Embodied experimentation and learnability for autonomous scientific discovery. AI for Nuclear and Fusion AI4Science methods for nuclear and fusion-oriented scientific discovery. MAFENN Multi-agent feedback enabled neural networks for intelligent communications.