PAND：面向提示的邻域蒸馏用于轻量级细粒度视觉分类
PAND: Prompt-Aware Neighborhood Distillation for Lightweight Fine-Grained Visual Classification
AI总结：
提出PAND框架，通过提示感知语义校准和邻域感知结构蒸馏，将大型视觉语言模型知识迁移至轻量网络，在细粒度分类任务上超越现有方法。
摘要：
在细粒度视觉分类（FGVC）中，从大型视觉语言模型（VLM）中蒸馏知识到轻量级网络至关重要但具有挑战性，原因是依赖于固定提示和全局对齐。为解决此问题，我们提出PAND（提示感知邻域蒸馏），一个两阶段框架，将语义校准与结构迁移解耦。首先，我们引入提示感知语义校准以生成自适应语义锚点。其次，我们提出邻域感知结构蒸馏策略以约束学生的局部决策结构。PAND在四个FGVC基准上持续优于现有方法。值得注意的是，我们的ResNet-18学生在CUB-200上达到76.09%的准确率，超过强基线VL2Lite 3.4%。代码可在https://github.com/LLLVTA/PAND获取。
英文摘要：
Distilling knowledge from large Vision-Language Models (VLMs) into lightweight networks is crucial yet challenging in Fine-Grained Visual Classification (FGVC), due to the reliance on fixed prompts and global alignment. To address this, we propose PAND (Prompt-Aware Neighborhood Distillation), a two-stage framework that decouples semantic calibration from structural transfer. First, we incorporate Prompt-Aware Semantic Calibration to generate adaptive semantic anchors. Second, we introduce a neighborhood-aware structural distillation strategy to constrain the student's local decision structure. PAND consistently outperforms state-of-the-art methods on four FGVC benchmarks. Notably, our ResNet-18 student achieves 76.09% accuracy on CUB-200, surpassing the strong baseline VL2Lite by 3.4%. Code is available at https://github.com/LLLVTA/PAND.
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。
