School of Humanities and Social Science Division of Social Science 230 The Political Economy of Conflict and Elections Supervisor: David James HENDRY / SOSC Student: WONG Ming Pan / GCS-SOSC Course: UROP 3200, Fall Previous literature suggests that economic grievances were a significant factor in civil conflicts; while such relationships have failed to be proven nationally, they are significant at the subnational level. Meanwhile, most theoretical arguments for grievance-conflict relationships involve an assumption that various political actors perceive those economic hardships. Hence, this project aimed to test the validity of subjective grievances and conflict relationships at the subnational level. This progress report is a follow-up report on the matchable terms in geographical information concerning Afrobarometer surveys, the IPUMS dataset, and the GIS information matches overall. It will also report potential issues of languages, directional naming strategies, and inconsistencies that would need further investigation to harmonize the dataset for reference in conducting MrP analysis in the future. Explainable AI from Cognitive Science Perspectives Supervisor: HSIAO Janet Hui-wen / SOSC Student: DUAN Qichun / QSA Course: UROP 1100, Spring UROP 2100, Summer This report presents findings from 226 studies evaluating the performance of large language models (LLMs) in human-centric scenarios, which focus on human-referenced and human-LLM interaction benchmarks. While assessing each type of benchmark individually, a key finding is that LLMs’ standalone performance does not always translate to effective human collaboration, particularly in dynamic settings like elderly care, in which conversational fluency and emotional sensitivity are essential. Additionally, the analysis highlights the growing number of research on LLMs’ ability in medical support, conversational dialogue, and writing assistance. The report proposes future research directions to develop comprehensive evaluation frameworks and enhance LLM interactive capabilities under caregiving scenarios. Corporate China: Structure and Change (II) Supervisor: LIN Yimin / SOSC Student: XIONG Zhijie / GCS-SOSC Course: UROP 1100, Fall As an extension of a project under the same title, offered at the 1000 level during the summer of 2024, this project continuously explored the multifaceted roles China’s top 500 enterprises played in China’s economic development and anticipated the economic challenges that may arise. By using VBA and python skills, a more holistic comparison between different enterprise ranking lists is conducted. To be more specific, we integrated the ownership structure of every Top 500 company for over 20 years. Then, we try to comprise the ranking differences of the same company on different lists, as well as the ranking differences of companies in different industries and ownership structures on various lists, in order to analyze and verify the ranking criteria of each list for companies in different industries and ownership structures. This is of great significance in eliminating potential biases in various lists.
RkJQdWJsaXNoZXIy NDk5Njg=