Preprint
Aug 2026
EVEREST:Endogenous Vision-Language Reinforcement Reasoning Exploration for Urban Socio-Semantic Segmentation
The EVEREST model, which adopts an egocentric exploration strategy that enables the model to actively investigate boundary cues and perform self-correction, achieves optimal performance on all metrics in the real world urban socio-semantic dataset, demonstrating the superiority of the model.
Qi-Xiu Li, Zhong-Zhi He, Xiang Zhu et al.
· 0 citations