Benchmarking Spurious Bias in Few-Shot Image Classifiers

TL;DR

This paper introduces FewSTAB, a systematic benchmark framework that automatically evaluates the robustness of few-shot image classifiers against spurious biases using attribute-based sample selection, addressing a key gap in current assessment methods.

Contribution

The paper presents FewSTAB, a novel automatic benchmarking system for quantifying spurious bias robustness in few-shot classifiers, utilizing attribute-based sample selection with pre-trained models.

Findings

FewSTAB effectively benchmarks spurious bias across multiple datasets.

It reveals varying degrees of robustness among different few-shot learning methods.

The framework guides the development of more robust few-shot classifiers.

Abstract

Few-shot image classifiers are designed to recognize and classify new data with minimal supervision and limited data but often show reliance on spurious correlations between classes and spurious attributes, known as spurious bias. Spurious correlations commonly hold in certain samples and few-shot classifiers can suffer from spurious bias induced from them. There is an absence of an automatic benchmarking system to assess the robustness of few-shot classifiers against spurious bias. In this paper, we propose a systematic and rigorous benchmark framework, termed FewSTAB, to fairly demonstrate and quantify varied degrees of robustness of few-shot classifiers to spurious bias. FewSTAB creates few-shot evaluation tasks with biased attributes so that using them for predictions can demonstrate poor performance. To construct these tasks, we propose attribute-based sample selection strategies based on a pre-trained vision-language model, eliminating the need for manual dataset curation. This allows FewSTAB to automatically benchmark spurious bias using any existing test data. FewSTAB offers evaluation results in a new dimension along with a new design guideline for building robust classifiers. Moreover, it can benchmark spurious bias in varied degrees and enable designs for varied degrees of robustness. Its effectiveness is demonstrated through experiments on ten few-shot learning methods across three datasets. We hope our framework can inspire new designs of robust few-shot classifiers. Our code is available at https://github.com/gtzheng/FewSTAB.