不用微调，直接用现成特征就能实现顶尖少样本学习效果。
Rethinking the Good Enough Embedding for Easy Few-Shot Learning
- 用冻结的DINOv2-L特征+kNN分类器，不训练直接推理。
- 在4个主流数据集上超越复杂元学习模型，性能领先。
- 通过PCA/ICA对特征流形优化，提升泛化能力，适合快速部署。
深度视觉识别正朝着通用表征方向演进。柏拉图表征假说指出，不同架构在大规模数据上训练后逐渐收敛至共享的“理想”嵌入空间。这引出关键问题：是否只需一个“好嵌入”就足够？本文利用这种收敛性，证明现成嵌入已足以应对复杂任务，无需任务特定微调。我们在少样本学习框架下验证该假设，提出一种无需反向传播的非参数化流程。通过在冻结的DINOv2-L特征上使用k-近邻分类器，并进行逐层分析以确定最优特征提取方式，进一步发现基于PCA和ICA的流形精炼具有有益的正则化作用。在四个主要基准测试中，该方法持续优于复杂的元学习算法，达到当前最佳性能。
原文摘要 · Abstract (English)
The field of deep visual recognition is undergoing a paradigm shift toward universal representations. The Platonic Representation Hypothesis suggests that diverse architectures trained on massive datasets are converging toward a shared, "ideal" latent space. This again raises a critical question: is a "Good Embedding All You Need?" In this paper, we leverage this convergence to demonstrate that off-the-shelf embeddings are inherently "good enough" for complex tasks, rendering intensive task-specific fine-tuning unnecessary. We explore this hypothesis within the few-shot learning framework, proposing a straightforward, non-parametric pipeline that entirely bypasses backpropagation. By utilizing a k-Nearest Neighbor classifier on frozen DINOv2-L features, we conduct a layer-wise characterization to identify an optimal feature extraction. We further demonstrate that manifold refinement via PCA and ICA provides a beneficial regularizing effect. Our results across four major benchmarks demonstrate that our approach consistently surpasses sophisticated meta-learning algorithms, achieving state-of-the-art performance.
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。