UCLIC Research Seminar Series
Social media platforms must detect and block a variety of unacceptable user-generated content, such such as adult or violent images. This detection task is difficult to automate due to high accuracy requirements, costs of errors, and nuanced rules for what is and is not acceptable. Consequently, platforms rely on a vast and largely invisible workforce of human moderators to filter such content. However, mounting evidence suggests that exposure to disturbing content can cause lasting psychological and emotional damage to some moderators. To mitigate such harm, my lab has been investigating various blur-based moderation interfaces for reducing exposure to disturbing content whilst preserving moderator ability to quickly and accurately flag it. We find that interactive blurring designs can reduce emotional impact without sacrificing moderation accuracy and speed. See our online demo at: http://ir.ischool.utexas.edu/CM/demo/. More broadly, promoting moderator wellness requires a more holistic approach, with wellness interventions spanning both programmatic and technological approaches. In an interdisciplinary project, involving health professionals who work with moderators, we review best practices, important directions for future work, and the need for greater academic-industry collaboration.
Link to talk recording : https://youtu.be/-LYyYmOInm8
Matthew Lease is an Amazon Scholar and an Associate Professor in the School of Information at the University of Texas at Austin. He is a faculty leader of UT Austin's Good Systems (http://goodsystems.utexas.edu/) Grand Challenge initiative to design responsible AI technologies. Lease's research on information retrieval spans search interfaces to algorithms, with a current focus on online fact-checking. In the field of human computation and crowdsourcing, Lease creates human-labeled datasets to train/test AI systems, and he designs human-in-the-loop systems to deliver "last-mile" capabilities where AI falls short. Lease has received several paper awards, as well as three early career awards from US funding agencies. For more information, please visit his homepage: https://www.ischool.utexas.edu/~ml/.