首个遥感视觉语言模型少样本适配基准，揭示模型表现差异。
Few-Shot Adaptation Benchmark for Remote Sensing Vision-Language Models
- 构建首个系统性少样本适配评估框架，覆盖10个遥感数据集。
- 三款先进模型在相同零样本性能下，少样本表现差异显著。
- 开源代码支持未来研究，适合遥感多模态模型开发者使用。
遥感视觉语言模型（RSVLMs）得益于大规模预训练，在多种任务上展现出强大的零样本性能。然而，其在低数据场景下的泛化能力，如少样本学习，仍缺乏充分探索。本文首次提出针对RSVLMs的少样本适配结构化基准。我们在十个遥感场景分类数据集上，对三种主流RSVLMs应用五种常见少样本适配策略进行综合实验。结果表明，尽管模型零样本性能相近，其少样本适应行为却存在显著差异，部分模型天生更易适配。现有方法表现波动大，无明确优劣，凸显开发更具鲁棒性的遥感专用少样本适配方法的必要性。为促进后续研究，我们提供可复现的基准框架与开源代码，支持在少样本条件下系统评估RSVLMs。源码已公开于GitHub：https://github.com/elkhouryk/fewshot_RSVLMs。
原文摘要 · Abstract (English)
Remote Sensing Vision-Language Models (RSVLMs) have shown remarkable potential thanks to large-scale pretraining, achieving strong zero-shot performance on various tasks. However, their ability to generalize in low-data regimes, such as few-shot learning, remains insufficiently explored. In this work, we present the first structured benchmark for evaluating few-shot adaptation methods on RSVLMs. We conduct comprehensive experiments across ten remote sensing scene classification datasets, applying five widely used few-shot adaptation strategies to three state-of-the-art RSVLMs with varying backbones. Our findings reveal that models with similar zero-shot performance can exhibit markedly different behavior under few-shot adaptation, with some RSVLMs being inherently more amenable to such adaptation than others. The variability of performance and the absence of a clear winner among existing methods highlight the need for the development of more robust methods for few-shot adaptation tailored to RS. To facilitate future research, we provide a reproducible benchmarking framework and open-source code to systematically evaluate RSVLMs under few-shot conditions. The source code is publicly available on Github: https://github.com/elkhouryk/fewshot_RSVLMs
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。