讲座时间：2025年5月9日 10:00-11:30
讲座嘉宾：香港大学经管学院 王大维
讲座主持：yl23411永利 吴苏青
讲座地点：yl23411永利会议室A523
讲座标题（Title）：
Human versus Large-Language Models Creativity: A Large-Scale Comparison of The Divergent Thinking Ability in Humans and Large-Language Models
人类vs大语言模型的创造力：人类和大语言模型的发散思维能力的大规模比较
讲座摘要（Abstract）:
This talk presents findings from a large-scale empirical study comparing the divergent creativity of humans and large language models (LLMs), including state-of-the-art systems like GPT-4 and DeepSeek-R1. Using the Divergent Association Task (DAT), a validated and algorithmically scored creativity measure, we analyzed over 10,000 human responses and more than 200,000 model-generated outputs. The results show that, on average, human creativity scores are marginally higher than those of LLMs. However, humans exhibit significantly greater variability and higher levels of originality at the upper tail of the distribution—an area where LLMs consistently underperform. These findings highlight fundamental differences in the generative processes of humans and LLMs.
本次讲座将介绍一项大规模实证研究，系统比较人类与大型语言模型（LLMs，如 GPT-4、DeepSeek-R1 等）的发散性创造力表现。研究采用经过验证、基于算法评分的创造力测评工具——Divergent Association Task（DAT），分析了约10,000名人类参与者和超过200,000条模型生成的回应。结果发现，人类在平均创造力得分上略高于 LLMs，但在人类样本中表现出更大的个体差异，尤其是在创造力分布的高端，展现出更高的独创性，而这一点是目前所有测试模型都无法企及的。研究揭示了人机在创造力生成机制上的根本性差异。
嘉宾简介（Bio）:
Professor Dawei Wang is currently an Assistant Professor at The University of Hong Kong Business School and serves as an external faculty member at the Northwestern Institute on Complex Systems (NICO), Northwestern University. He received his Ph.D. in Management and Organizations from the Kellogg School of Management at Northwestern University.
Professor Wang’s research focuses on human-AI collaboration, exploring how AI can independently or cooperatively support human performance in domains such as creativity, healthcare, and scientific research. His work integrates experimental methods and computational approaches to examine the mechanisms of effective human-AI teaming and to identify the conditions under which AI can complement human judgment, innovation, and problem-solving capabilities.
王大维教授现任香港大学经济及工商yl23411永利助理教授，同时兼任美国西北大学复杂系统研究所（NICO）的外聘教授。他于美国西北大学凯洛格商学院取得管理与组织学博士学位。
王教授的研究聚焦于人机协作，探讨人工智能在创造力、医学、科学研究等领域中，如何以独立或协作的方式有效支持人类的任务执行。他通过实验研究与计算建模相结合的方法，系统性分析人机协作的机制，深入理解人工智能在辅助人类判断、创新与问题解决过程中的优势与局限。
Personal page: https://www.hkubs.hku.hk/people/dawei-wang/
欢迎广大感兴趣的师生积极参与！
