Training-free enhancement of satellite remote sensing VLMs via Geo-Contrastive Decoding

Abstract Vision language models (VLMs) have opened new avenues for satellite remote sensing image analysis and have shown promise across multiple tasks. However, in the absence of a remote sensing–oriented general VLM, existing approaches rely on retraining generic VLMs with remote sensing datasets...

Fuld beskrivelse

Bibliografiske detaljer

Principais autores:	Zixuan Shangguan, Jingrui Zhang, Ke Xing, Zhuohao Gong, Yong Zhang, Yimeng Xu, Feng Liang
Format:	Artigo
Sprog:	Inglês
Udgivet:	SpringerOpen 2026-03-01
Serier:	Journal of Cloud Computing: Advances, Systems and Applications
Fag:	Vision language models Satellite Remote sensing Contrastive decoding
Online adgang:	https://doi.org/10.1186/s13677-026-00885-7
Tags:	Tilføj Tag Ingen Tags, Vær først til at tagge denne postø!