Papers › Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and...

Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference

15 Apr 2022CVPR 2022 1arXiv:2204.07305archive 2025-07-28

Shell Xu Hu, Da Li, Jan Stühmer, Minyoung Kim, Timothy M. Hospedales

Few-shot learning (FSL) is an important and topical problem in computer vision that has motivated extensive research into numerous methods spanning from sophisticated meta-learning methods to simple transfer learning baselines. We seek to push the limits of a simple-but-effective pipeline for more realistic and practical settings of few-shot image classification. To this end, we explore few-shot learning from the perspective of neural network architecture, as well as a three stage pipeline of network updates under different data supplies, where unsupervised external data is considered for pre-training, base categories are used to simulate few-shot tasks for meta-training, and the scarcely labelled data of an novel task is taken for fine-tuning. We investigate questions such as: (1) How pre-training on external data benefits FSL? (2) How state-of-the-art transformer architectures can be exploited? and (3) How fine-tuning mitigates domain shift? Ultimately, we show that a simple transformer-based pipeline yields surprisingly good performance on standard benchmarks such as Mini-ImageNet, CIFAR-FS, CDFSL and Meta-Dataset. Our code and demo are available at https://hushell.github.io/pmf.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2204.07305")

Code

Syntology Ran 14 of 19 code samples harvested from 1 repository linked to this paper; 5 have no recorded run. Of those that ran: 4 ran · our draft was wrong; 2 ran · fixture could not drive it; 8 ran with no contract checked.

By repository: official repository: 19 samples from 1 repository, 14 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

hushell/pmf_cvpr22 officialmentioned in papermentioned on GitHubpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

19 samples harvested; 14 ran; 0 honoured the contract we drafted; 5 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

4ran · our draft was wrong
2ran · fixture could not drive it
8ran
5unverified

Licence: 0 of the 19 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from hushell/pmf_cvpr22. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

np2th hushell/pmf_cvpr22/models/resnet_v2.py official repository ran · fixture could not drive it Apache-2.0 (permissive) · 7d8eaf5537f699e4 · report
swish hushell/pmf_cvpr22/models/vit_google.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 0f786c407fb1ee4c · report
DiffAugment hushell/pmf_cvpr22/models/utils.py official repository ran Apache-2.0 (permissive) · 98235a49e66df4c8 · report
build_transform hushell/pmf_cvpr22/datasets/cifar_fs_elite.py official repository ran Apache-2.0 (permissive) · f0f38cfab1ecddd0 · report
conv1x1 hushell/pmf_cvpr22/models/resnet_v2.py official repository ran · our draft was wrong Apache-2.0 (permissive) · b0f916df39e78ae5 · report
conv3x3 hushell/pmf_cvpr22/models/resnet_v2.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 4e5bf13dbdc4f008 · report
dataset_setting hushell/pmf_cvpr22/datasets/mini_imagenet.py official repository ran fingerprinted Apache-2.0 (permissive) · 1511126d79e9b40b · report
dataset_setting hushell/pmf_cvpr22/datasets/cifar_fs.py official repository ran fingerprinted Apache-2.0 (permissive) · b1b53caaa091d2ba · report
dataset_setting hushell/pmf_cvpr22/datasets/cifar_fs_elite.py official repository ran fingerprinted Apache-2.0 (permissive) · 196d8eff1ed207d4 · report
drop_path hushell/pmf_cvpr22/models/vision_transformer.py official repository ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 55120f2026b56aa2 · report
entropy_loss hushell/pmf_cvpr22/models/deploy.py official repository ran Apache-2.0 (permissive) · af188f96f841e377 · report
rand_brightness hushell/pmf_cvpr22/models/utils.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 011230b2b9b8fb6f · report
random_hflip hushell/pmf_cvpr22/models/utils.py official repository ran fingerprinted Apache-2.0 (permissive) · e3e99bdecc02eb33 · report
unique_indices hushell/pmf_cvpr22/models/deploy.py official repository ran fingerprinted Apache-2.0 (permissive) · b1abb876763549d7 · report
beit_base_patch16_224 hushell/pmf_cvpr22/models/beit.py official repository unverified Apache-2.0 (permissive) · 051c35752a444b56 · report
beit_base_patch16_384 hushell/pmf_cvpr22/models/beit.py official repository unverified Apache-2.0 (permissive) · 657ea09a85528243 · report
beit_large_patch16_224 hushell/pmf_cvpr22/models/beit.py official repository unverified Apache-2.0 (permissive) · 7ef377ba295aa2db · report
vit_small hushell/pmf_cvpr22/models/vision_transformer.py official repository unverified Apache-2.0 (permissive) · 14af5cea1b69d0a8 · report
vit_tiny hushell/pmf_cvpr22/models/vision_transformer.py official repository unverified Apache-2.0 (permissive) · bd29518a98e56cf5 · report

Tasks

Few-Shot Image ClassificationFew-Shot LearningImage ClassificationMeta-LearningTransfer Learningimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Few-Shot Image Classification CIFAR-FS 5-way (1-shot) P>M>F (P=DINO-ViT-base, M=ProtoNet) Accuracy 84.3 #10 of 38 Archive leaderboard report
Few-Shot Image Classification CIFAR-FS 5-way (5-shot) P>M>F (P=DINO-ViT-base, M=ProtoNet) Accuracy 92.2 #4 of 39 Archive leaderboard report
Few-Shot Image Classification Meta-Dataset P>M>F (P=DINO-ViT-base, M=ProtoNet) Accuracy 84.75 #2 of 22 Archive leaderboard report
Few-Shot Image Classification Mini-Imagenet 5-way (1-shot) P>M>F (P=DINO-ViT-base, M=ProtoNet) Accuracy 95.3 #3 of 105 Archive leaderboard report
Few-Shot Image Classification Mini-Imagenet 5-way (5-shot) P>M>F (P=DINO-ViT-base, M=ProtoNet) Accuracy 98.4 #3 of 95 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AttentionDINODense ConnectionsLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections