报告时间：2025年9月27日上午9:00-12:00
报告地点：计算机学院B404会议室
报告一：Investigating the Impact of Online Community Involvement on Safety Practices and Perceived Risks Among People Who Use Drugs
报告人：王也 助理教授
报告人单位：澳门大学
报告人简介：王也，澳门大学电脑及资讯科学系、金融及商业经济系双聘助理教授，博导。在苏黎世联邦理工学院获得博士和硕士学位，在北京大学获得学士学位。主要从事信息安全、人机交互、金融科技等方面的研究工作。近年来在JFE、S&P、USENIX Security、CCS、NDSS、CHI、CSCW、WWW、WINE、IJCAI等国际期刊与会议发表学术论文30余篇。主持/共同主持国家科技部重点研发计划、澳门科学技术发展基金面上项目等科研项目。
报告摘要：Substance abuse is prevalent in contemporary society, posing significant safety risks to people who use drugs (PWUD). The advent of online communities is gradually altering PWUD's drug use behaviors and, as a consequence, their safety practices. Yet, the precise impacts and challenges these changes introduce to PWUD's safety have scarcely been explored. We conducted 19 semi-structured interviews with PWUD and analyzed content from 29 drug-related subreddits to examine how their involvement in online communities influences their safety practices. Our findings reveal that online communities have afforded PWUD broader access to information and community engagement, facilitating emotional support and better management of their drug use. Nonetheless, it has also introduced challenges such as the spread of misinformation, the promotion of risky behaviors, and the risk of deception. We further discuss the challenges PWUD face transitioning to online communities, and offer design implications to enhance their safety.
报告二：Beyond Acceptance Models: Configurational Pathways of Generative AI Adoption in Student Error Analysis
报告人：林鑫 助理教授
报告人单位：澳门大学
报告人简介：林鑫，现任澳门大学教育学院助理教授、博士生导师，毕业于美国德克萨斯大学奥斯汀分校。她的研究聚焦于学习科学与人工智能的结合，主要探讨数学学习困难学生的认知特点、智能干预策略，以及生成式人工智能在教育与特殊教育中的应用。近年来，林博士以第一作者或通讯作者身份在《Review of Educational Research》《Educational Psychology Review》《Child Development》等国际顶级SSCI期刊及CCF A类会议发表论文四十余篇。她的研究广泛采用荟萃结构方程建模（MetaSEM）、实验设计和人机交互方法，系统揭示数学认知发展的关键机制，并探索人工智能技术在促进个性化学习和教育公平方面的潜力。
报告摘要：Generative AI (GAI) tools are becoming common in education, but it is still unclear why and how people choose to use them for complex tasks. This study looks at pre-service mathematics teachers using GAI to analyze student mistakes—a challenging activity that requires deep reasoning. We surveyed 213 teachers in China and applied fuzzy-set qualitative comparative analysis (fsQCA) to explore six factors: beliefs, self-efficacy, error-analysis experience, and perceptions of usefulness, ease of use, and exploratory disposition. The analysis revealed five equally effective adoption pathways—profession-grounded, exploration-enhanced, convenience-driven, belief–experience, and value–exploration models. Follow-up interviews further showed how pedagogical orientation, technological perceptions, and exploratory mindset combine to drive adoption. Unlike traditional linear acceptance models, which assume adoption can be explained by a single dominant predictor, our findings demonstrate that adoption emerges through multiple interacting configurations. For HCI, this underscores that GAI adoption is not one-size-fits-all, and systems should be designed to accommodate diverse pathways, supporting exploration, sensemaking, and alignment with users’ goals and values.
邀请人：程大钊、梁黄黄
