Multimodal Agents: From Automation toward Open-Ended Self-Improvement Mingchen Zhuge, Ph.D. Student, Computer Science May 9, 17:30 - 19:30 B4 R5220; Zoom Meeting 91489077683 AI agents coding Multi-agent systems world models recursive self-improvement LLM Deep Reinforcement Learning This thesis presents practical methodologies for building scalable multimodal agents that move from narrow automation toward open-ended self-improvement.
Learning to Identify and Exploit Neural Network Dynamics in Multi-Step Inference Haozhe Liu, Ph.D. Student, Computer Science Mar 9, 13:30 - 15:30 B4 L5 R5220; Zoom Meeting 95866424218 This dissertation studies the temporal dynamics of multi-step inference and reveals that contributions across steps are sparse and uneven.
