Beyond the Explicit: Benchmarking and Detecting Implicit Harm in Short Videoswith Chenyang Li, Qifan Yu, Saihui Hou1 versionW_enucnhqtv1 · current