Are We Really Doing Few-Shot Learning? A Critical Examination of Pre-Training Assumptions

📅 2026-09-09
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
本文通过对比不同预训练方法,探讨了少样本学习评估协议是否真正反映了低数据学习情况,提出了基于描述符的源选择策略。
📝 Abstract
Few-shot learning is commonly evaluated under protocols that pre-train a model on a large auxiliary set whose classes are disjoint from the target episodes yet drawn from the same visual domain. This paper examines whether such protocols truly reflect low-data learning. We systematically compare no pre-training, class-disjoint in-domain pre-training, supervised out-of-domain pre-training, and label-free out-of-domain pre-training across eight datasets, three few-shot architectures, and multiple way-shot settings. Our results show that class disjointness alone is insufficient to remove the influence of target-domain data. In-domain pre-training improves over no pre-training by 33.41 percentage points on average, whereas supervised out-of-domain pre-training yields 23.75 percentage points, revealing a 9.66-point optimistic bias associated with domain overlap. Although out-of-domain pre-training is more realistic in applications where target-domain data are scarce, its effectiveness depends strongly on the compatibility between source and target domains. We further show that labeled source data are not strictly required, with an augmentation-based label-free strategy reaching an average gain of 27.71 percentage points and closely matching supervised out-of-domain pre-training at 27.97 percentage points. Finally, we introduce a descriptor-based source-selection strategy that estimates source-domain suitability before pre-training, reaching a median gap of only 1.37 percentage points to oracle selection. These findings highlight the need to move beyond in-domain pre-training as the default few-shot evaluation protocol, since it can overestimate performance in realistic scenarios where target-domain data are scarce.
Problem

Research questions and friction points this paper is trying to address.

few-shot learning
pre-training
low-data learning
domain overlap
Innovation

Methods, ideas, or system contributions that make the work stand out.

class-disjoint in-domain pre-training
label-free out-of-domain pre-training
source-selection strategy
🔎 Similar Papers
No similar papers found.
A
Alejandro Galan-Cuenca
University Institute for Computer Research, University of Alicante, Alicante, Spain
M
Marcelo Saval-Calvo
University Institute for Computer Research, University of Alicante, Alicante, Spain
Antonio Javier Gallego
Antonio Javier Gallego
University Institute for Computer Research, University of Alicante, Alicante, Spain