arXiv 2606.207022026-06-23cs.CVcs.AIcs.LG

超越模板：通过元提示重新审视零样本遥感

Beyond Templates: Revisiting Zero-Shot Remote Sensing through Meta-Prompting

Remote Sensing Lab, National Technical University of Athens（遥感实验室，国家技术大学雅典分校）；Remote Sensing Lab, Department of Engineering and Sciences, National Technical University of Athens（遥感实验室，工程与科学系，国家技术大学雅典分校）；Universitas Mercatorum（默卡托大学）

Eirini Baltzi, Dionysis Christopoulos, Sotiris Spanos, Valsamis Ntouskos, Konstantinos Karantzalos

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

研究通过元提示框架（MPVR）在17种视觉语言模型和12个遥感数据集上探索零样本性能，发现语义丰富的LLM生成描述可能引入噪声，而轻量级查询嵌入校准能稳定提升分类和检索性能。

摘要：

视觉语言模型（VLM）在零样本地球观测（EO）下游任务中引发了日益增长的兴趣，而遥感适配模型进一步提升了性能。我们在元提示视觉识别（MPVR）框架下，对17种VLM变体和12个遥感（RS）数据集进行了考察，结果表明零样本性能对文本设计选择高度敏感，从用于引导LLM生成类别描述的元提示到描述本身都是如此。我们探讨了为什么语义丰富的LLM生成的类别描述并未比简单的领域适配CLIP风格描述带来一致的提升。虽然LLM描述在语义上更具表现力，但它们也可能在文本嵌入空间中引入噪声，从而降低下游任务的鲁棒性。我们通过白化CLIP特征空间中的文本对数似然分析支持了这一观察，比较了LLM生成描述和基于模板的描述。基于这一发现，我们研究了查询嵌入校准，并表明查询空间的轻量级校准在零样本分类和检索中始终能带来显著的改进。总体而言，我们的结果为语义丰富性与鲁棒性之间的权衡提供了实用见解，并将嵌入校准确定为改进零样本遥感性能的简单有效工具。

英文摘要：

Vision-language models (VLMs) have sparked growing interest in zero-shot Earth Observation (EO) downstream tasks, with further gains enabled by remote-sensing-adapted models. We examine this setting across 17 VLM variants and 12 remote sensing (RS) datasets under Meta-Prompting for Visual Recognition (MPVR), and show that zero-shot performance remains highly sensitive to textual design choices, from the meta-prompts used to guide the LLM in generating class descriptions to the descriptions themselves. We explore why semantically rich LLM-generated class descriptions do not translate into consistent gains over simple domain-adapted CLIP-style descriptions. While LLM descriptions are more semantically expressive, they can also introduce noise in the text embedding space, reducing robustness in downstream tasks. We support this observation through a text log-likelihood analysis in the whitened CLIP feature space, comparing LLM-generated and template-based descriptions. Building on this finding, we study query embedding calibration and show that lightweight calibration of the query space consistently yields strong improvements in zero-shot classification and retrieval. Overall, our results provide practical insight into the trade-off between semantic richness and robustness, and identify embedding calibration as a simple and effective tool for improving zero-shot remote sensing performance.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。