CAILA: 面向组合式零样本学习的概念感知层内适配器
CAILA: Concept-Aware Intra-Layer Adapters for Compositional Zero-Shot Learning
AI总结：
针对组合式零样本学习中现有CLIP类方法未挖掘内部层语义概念的问题，本文提出概念感知层内适配器CAILA，在CLIP各编码层插入可提取三类概念特征的参数高效适配器，在四个基准数据集上取得最优性能。
摘要：
本文研究组合式零样本学习（Compositional Zero-Shot Learning, CZSL）问题，即利用预存概念识别全新的属性-对象组合。近期研究人员聚焦于应用具备强泛化能力的大规模视觉语言预训练（Vision-Language Pre-trained, VLP）模型（如CLIP）。但这类方法将预训练模型视为黑盒，仅关注CLIP的前后置操作，并未从本质上挖掘CLIP内部各层之间的语义概念。我们提出深入模型架构，在CLIP的每个编码器层中插入适配器——一种已在大语言模型中被验证有效的参数高效技术。我们进一步为适配器赋予概念感知能力，从而可提取“对象”“属性”和“组合”的特定概念特征。我们在四个主流CZSL数据集（MIT-States、C-GQA、UT-Zappos和VAW-CZSL）上评估了所提方法，结果显示其在所有数据集上的性能均优于现有方法，达到当前最优水平。
英文摘要：
In this paper, we study the problem of Compositional Zero-Shot Learning (CZSL), which is to recognize novel attribute-object combinations with pre-existing concepts. Recent researchers focus on applying large-scale Vision-Language Pre-trained (VLP) models like CLIP with strong generalization ability. However, these methods treat the pre-trained model as a black box and focus on pre- and post-CLIP operations, which do not inherently mine the semantic concept between the layers inside CLIP. We propose to dive deep into the architecture and insert adapters, a parameter-efficient technique proven to be effective among large language models, into each CLIP encoder layer. We further equip adapters with concept awareness so that concept-specific features of "object", "attribute", and "composition" can be extracted. We assess our method on four popular CZSL datasets, MIT-States, C-GQA, UT-Zappos, and VAW-CZSL, which shows state-of-the-art performance compared to existing methods on all of them.
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。
