Talk: Autonomous AI agents
Shuyan Zhou is giving a talk on autonomous AI agents on Tuesday, Nov. 7.
Date: Tuesday, Nov. 7 from 10:30AM to 11:30AM
Location: Cockins Hall 240 (1958 Neil Ave, Columbus)
Abstract: For years, my dream has been to create autonomous AI agents capable of carrying out tedious tasks, allowing me to focus on more creative and exciting tasks. Modern AI models, especially large language models (LLMs) like ChatGPT, have suddenly brought us much closer to achieving such AI agents. But, has my dream already come true? In 2020, I started research on AI agents, through my experience, I have recognized LLMs as a crucial yet early step in the broader landscape of AI agent research. While LLMs show promising results on well-specified tasks such as drafting and sending emails or answering everyday questions, they struggle with more complex tasks such as coordinating a meeting for multiple participants. Such tasks often span a long horizon and require a wealth of procedural knowledge and dynamic interaction with the environment. Current LLMs cannot perform such challenging tasks with consistent accuracy and even may pose safety risks due to their unpredictable behavior. In this talk, I will present three thrusts in my research toward this dream: the formalism of planning and acting, the learning of open-domain procedural knowledge, and the evaluation of AI agents.
Bio: Shuyan Zhou (https://shuyanzhou.com) is a final-year Ph.D. student from the Language Technologies Institute at Carnegie Mellon University, advised by Prof. Graham Neubig. She studies autonomous AI agents and she aims to create AI agents that could perform tedious tasks so that everyone can focus on more creative and interesting tasks. Before joining CMU, she received her B.S. degree from the Department of Computer Science and Technology at Harbin Institute of Technology.