# Are Few-Shot Learning Benchmarks too Simple ? Solving them without Task Supervision at Test-Time

arXiv: https://arxiv.org/abs/1902.08605 · HTML: https://mlsanity.com/paper/1902.08605/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Feb 2019.

## Built on by (most cited first, 4 of 4)

- [Learning from Few Examples: A Summary of Approaches to Few-Shot Learning](https://mlsanity.com/paper/2203.04291/) (2022)
- [A Survey of Few-Shot Learning for Biomedical Time Series](https://mlsanity.com/paper/2405.02485/) (2024)
- [Online Unsupervised Learning of Visual Representations and Categories](https://mlsanity.com/paper/2109.05675/) (2021)
- [Few-Example Clustering via Contrastive Learning](https://mlsanity.com/paper/2207.04050/) (2022)

## Abstract (arXiv)

We show that several popular few-shot learning benchmarks can be solved with varying degrees of success without using support set Labels at Test-time (LT). To this end, we introduce a new baseline called Centroid Networks, a modification of Prototypical Networks in which the support set labels are hidden from the method at test-time and have to be recovered through clustering. A benchmark that can be solved perfectly without LT does not require proper task adaptation and is therefore inadequate for evaluating few-shot methods. In practice, most benchmarks cannot be solved perfectly without LT, but running our baseline on any new combinations of architectures and datasets gives insights on the baseline performance to be expected from leveraging a good representation, before any adaptation to the test-time labels.