LLM Agents: From Language to Action

In recent years, the emergence of Large Language Models (LLMs) has revolutionized artificial intelligence, enabling machines to understand and generate human-like text with unprecedented capabilities. However, the true potential of LLMs extends far beyond language generation—when combined with intelligent agents, they bridge the critical gap between understanding and action. This seminar will explore the transformative synergy between LLMs and Autonomous Agents. Prof. Huang will investigate cutting-edge research, including advancements driven by LLM Agents that enable systems to interpret natural language instructions and execute complex tasks, such as programming, web search, information analysis, and comprehensive report generation, in real-world dynamic environments. Tools will be demonstrated during the session:

1. Auto-Deep-Research: https://github.com/HKUDS/Auto-Deep-Research
2. AutoAgent: https://github.com/HKUDS/AutoAgent

After completing this session, participants will be able to

1. Understand the fundamental capabilities of Large Language Models (LLMs) and how they extend beyond basic text generation to enable complex reasoning and task execution.
2. Identify and analyze the synergistic relationship between LLMs and Autonomous Agents, and how this combination creates systems capable of bridging understanding with real-world actions.
3. Evaluate cutting-edge applications of LLM Agents across diverse domains, including programming, web search, information analysis, and report generation in dynamic environments.