1 paper
Zhirui Fang, Ming Yang, Weishuai Zeng +5
We explore leveraging large multi-modal models (LMMs) and text2image models to build a more general embodied agent. LMMs excel in planning long-horizon tasks over symbolic abstract…