主题：Unravel unraveling
时间：2026年9月11日 10:00-11:30
地点：管理学院 A823
主讲人：张琪，香港中文大学 助理教授
主持人：吴志岩，浙江大学管理学院 教授
嘉宾简介（Bio）:
Prof. Zhang Qi is an Assistant Professor of Accounting at the Chinese University of Hong Kong (CUHK). His research is primarily focused on corporate disclosure, machine learning, and corporate sustainability. He received his PhD and MSc degrees from Frankfurt School of Finance and Management, and BA from the Central University of Finance and Economics. His work has been published in the Journal of Accounting Research.
摘要（Abstract）:
This study explores the ability of Large Language Models (LLMs) to analyze strategically framed human capital disclosures, where firms often engage in selective reporting that highlights strengths while withholding vulnerabilities. Through experimentation, we demonstrate that LLMs can effectively summarize explicit disclosures, but uncovering hidden risks requires carefully designs. One effective approach combines harness engineering with contextual business knowledge. By benchmarking these evaluations against Glassdoor employee reviews, we show that, with proper guidance, LLMs can navigate incomplete disclosures and identify omitted weaknesses. As regulatory delays in standardizing human capital disclosures persist, our findings underscore the potential of LLMs to transform transparency and accountability in corporate communications.
欢迎广大师生前来交流！
