作者位次 #8/9ACL 2026Poster· ACL Industry Track
UNIVID : Unified Vision-Language Model for Video Moderation
Kejuan Yang · Yizhuo Zhang · Mingyuan Du · Yue Zhang · Dixin Zheng · Kaili Zhao · Yang Xiao · Hanzhong Liang · Kenan Xiao
摘要
Global-scale video moderation faces a dual challenge: the need for fine-grained multimodal reasoning and the demand for interpretable outputs to support downstream enforcement. Traditional moderation systems often rely on fragmented black-box classifiers that are difficult to maintain and lack tr...