E2E-GMNER: End-to-End Generative Grounded Multimodal Named Entity Recognition

Meng Zhang, Jinzhong Ning, Xiaolong Wu +2

Grounded Multimodal Named Entity Recognition (GMNER) aims to jointly identify named entity mentions in text, predict their semantic types, and ground each entity to a corresponding…