BotaCLIP: Contrastive Learning for Botany-Aware Representation of Earth Observation Data

ArXi:2511.21194v2 Announce Type: replace Foundation models have nstrated a remarkable ability to learn rich, transferable representations across diverse modalities such as images, text, and audio. In modern machine learning pipelines, these representations often replace raw data as the primary input for downstream tasks. In this paper, we address the challenge of adapting a pre-trained foundation model to inject domain-specific knowledge, without re