基于在线视频平台的人工智能生成摘要设计：来自实地实验的证据
发布时间：2025-06-11
6月
11
|
时间和日期
|
2025-06-11 (星期三) 10:30 上午
-
12:00 下午
|
| 标题: | 基于在线视频平台的人工智能生成摘要设计：来自实地实验的证据 |
| 日期和时间: |
2025年6月11日（周三） 10:30-12:00 |
| 地点 | 综合教学楼D604阶梯教室 |
| 主讲人: |
高超越 中国科学技术大学 |
| 摘要: | AI-generated summaries condense lengthy text or video content into short and concise textual summaries, which can facilitate consumer search and reduce information overload. This research designs and evaluates two AI-generated summarization strategies for online videos: information-extractive AI-generated summaries (IAIGS), which present fact-based recaps, and suspense-inducing AI-generated summaries (SAIGS), which withhold key details to spark curiosity. To assess their impact on video consumption, we conducted a randomized field experiment on a video-sharing platform, focusing on two content genres: Science, which primarily addresses users’ instrumental information needs, and Humanities & History, which caters to users’ affective and curiosity-driven interests. We analyzed engagement metrics for 21,533 videos from 1,545 content creators over a six-week period. Our results show that IAIGS consistently reduced video views across both genres, driven by information substitution. In contrast, SAIGS had genre-specific effects: it increased engagement with Humanities & History content by sparking curiosity, but decreased engagement with Science content by obstructing information-seeking. A follow-up online experiment confirmed these patterns and shed light on the underlying mechanisms. Our study highlights the nuanced effects of AI summarization strategies on user engagement across content genres, emphasizing the importance of aligning summary design with user intent and content characteristics. Our findings provide valuable insights for designing AI-generated summaries to enhance user engagement and content consumption. |
| 主讲人简介: | 高超越博士是中国科学技术大学管理学院的特任副教授，研究方向为信息系统。他于哈尔滨工业大学获得管理学博士学位，并同时在香港城市大学资讯系统系获得博士学位。此外，他还取得了哈尔滨工业大学的管理学学士与工学学士学位、管理学辅修学位以及工学硕士学位。他的研究主要聚焦于区块链技术与大语言模型（LLMs）的实际应用，探讨其对用户行为的影响。在研究方法上，他主要采用二手数据分析，辅以实验室实验或实地实验以及设计科学方法。他的研究成果发表于《Journal of Operations Management》, 《Production and Operations Management》, 《Decision Support Systems》, 管理世界等多个国内外顶级学术期刊，并曾多次获得最佳论文奖及最佳论文提名奖（ICIS 2024、AIS SIGBIT 2024、CSWIM 2022、CSWIM 2023、PACIS 2023、POMS International Conference in China 2024）。 |
