发表机构
Faculty of Science and Technology; Norwegian University of Life Sciences; NMBU（科学与技术学院; 挪威生命科学大学; NMBU）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出无训练无图像的CAST框架，通过权重注入扩展预训练分类器至未见类别，其性能接近少样本适配方法，为零样本学习提供了新的无图像解决方案。
AI 中文摘要
大型预训练模型已成为现代机器学习系统的基础组件，但将这些模型适配到新类别通常需要目标分布的样本。然而在许多领域，这类数据无法获取。零样本学习（ZSL）通过依赖文本描述等辅助语义信息，允许在这种限制下进行识别。我们提出CAST（Closed-form Analytic Semantic Transfer，闭式解析语义迁移），这是一种无训练、无图像的框架，用于通过权重注入将预训练分类器扩展到先前未见过的类别。我们为CAST提供理论基础，并推导有限样本误差分解，识别出语义外推残差ρ_u。该残差是可计算的、与模型无关的度量，为数据集整理和基准设计提供了原则性标准。在标准零样本学习基准上的实验表明，CAST的性能与现有无图像方法相当或更优，且接近少样本适配方法的性能，同时既不需要迭代优化，也不需要目标分布的样本。
英文摘要
Large pre-trained models have become foundational components of modern machine learning systems. Yet adapting these models to novel categories typically requires examples from the target distribution. In many domains, however, such data are unavailable. Zero-shot learning (ZSL) permits recognition under these limitations through relying on auxiliary semantic information such as textual descriptions. We introduce CAST (Closed-form Analytic Semantic Transfer), a training-free, image-free framework for extending a pre-trained classifier to previously unseen classes through weight injection. We provide a theoretical foundation for CAST and derive a finite-sample error decomposition that identifies the \emph{semantic extrapolation residual} $ρ_u$. The residual is a computable, model-agnostic measure and provides a principled criterion for dataset curation and benchmark design. Experiments on standard zero-shot learning benchmarks demonstrate that CAST matches or exceeds existing image-free approaches and approaches the performance of few-shot adaptation methods, while requiring neither iterative optimization nor examples from the target distribution.
