Papers › Big Transfer (BiT): General Visual Representation Learning

Big Transfer (BiT): General Visual Representation Learning

24 Dec 2019ECCV 2020 8arXiv:1912.11370archive 2025-07-28

Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Joan Puigcerver, Jessica Yung, Sylvain Gelly, Neil Houlsby

Transfer of pre-trained representations improves sample efficiency and simplifies hyperparameter tuning when training deep neural networks for vision. We revisit the paradigm of pre-training on large supervised datasets and fine-tuning the model on a target task. We scale up pre-training, and propose a simple recipe that we call Big Transfer (BiT). By combining a few carefully selected components, and transferring using a simple heuristic, we achieve strong performance on over 20 datasets. BiT performs well across a surprisingly wide range of data regimes -- from 1 example per class to 1M total examples. BiT achieves 87.5% top-1 accuracy on ILSVRC-2012, 99.4% on CIFAR-10, and 76.3% on the 19 task Visual Task Adaptation Benchmark (VTAB). On small datasets, BiT attains 76.8% on ILSVRC-2012 with 10 examples per class, and 97.0% on CIFAR-10 with 10 examples per class. We conduct detailed analysis of the main components that lead to high transfer performance.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

No sample from the archive's official repository (google-research/big_transfer) is linked to this paper; the 10 samples below come from 1 community repository. Syntology harvested code from google-research/big_transfer for another paper.

google-research/big_transfer: Syntology's graph links these samples to Robust and Efficient Medical Imaging with Self-Supervision (arXiv:2205.09723; that paper's own run record: 20 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 1 violated, 15 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method.

Syntology 10 code samples harvested from 1 repository linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 10 samples from 1 repository, 9 ran (7 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

google-research/big_transfer officialmentioned on GitHubjaxApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Robust and Efficient Medical Imaging with Self-Supervision (arXiv:2205.09723; that paper's own run record: 20 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 1 violated, 15 with no contract checked; 4 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
SoojungYang/supervised_pretraining_GN_WS mentioned on GitHubtfSyntology: not harvested report
batsresearch/taglets mentioned on GitHubpytorchApache-2.0Syntology: 10 samples, 9 ran (2 instrument-failed) · 1 unverified report
bethgelab/InDomainGeneralizationBenchmark mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Visual Representation Learning Does Not Generalize Strongly Within the Same Domain (arXiv:2107.08221; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
sayakpaul/A-Barebones-Image-Retrieval-System mentioned on GitHubtfMITSyntology: not harvested report
sayakpaul/FunMatch-Distillation mentioned on GitHubtfSyntology: not harvested report
2024-MindSpore-1/Code2 mindsporeSyntology: not harvested report
MS-Mind/MS-Code-02 mindsporeApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to EdgeNeXt: Efficiently Amalgamated CNN-Transformer Architecture for Mobile Vision Applications (arXiv:2206.10589; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
hw666666666666/BigTransfer mindsporeNOASSERTIONSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

10 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
2ran · our draft was wrong
6ran
1unverified

Licence: 0 of the 10 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 batsresearch/taglets. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

batsresearch/taglets community (archive-listed) 10 samples, 9 ran (2 instrument-failed) · 1 unverified
Brier_Score_AMCL taglets/labelmodel/amcl_helper.py ran fingerprinted Apache-2.0 (permissive) · 00301e057c30f413 · report fixture: signature preds:ndarray/2/float · returned ndarray · level recorded 2026-09-24
Brier_loss_linear taglets/labelmodel/amcl_helper.py ran fingerprinted Apache-2.0 (permissive) · 081ac982970319a1 · report fixture: signature labels:ndarray/1/float, preds:ndarray/1/float · returned float64 · level recorded 2026-09-24
conv1x1 in 691 papers taglets/models/resnetv2.py ran · our draft was wrong Apache-2.0 (permissive) · ee80cf467065076b · report fixture: signature cin:int/0/int, cout:int/0/int, stride:int/0/int, bias:bool/0/bool · returned StdConv2d · level recorded 2026-09-22
conv3x3 in 1,376 papers taglets/models/resnetv2.py ran · our draft was wrong Apache-2.0 (permissive) · f01b8d3901f0b289 · report fixture: signature cin:int/0/int, cout:int/0/int, stride:int/0/int, groups:int/0/int, bias:bool/0/bool · returned StdConv2d · level recorded 2026-09-22
get_data in 223 papers taglets/labelmodel/amcl.py ran Apache-2.0 (permissive) · 174d041dba5d15c6 · report fixture: signature num:int/0/int · returned tuple · level recorded 2026-09-24
get_test_data in 10 papers taglets/labelmodel/amcl.py ran Apache-2.0 (permissive) · fbc44ea7cbcc39ee · report fixture: signature num:int/0/int · returned tuple · level recorded 2026-09-24
get_test_data in 10 papers taglets/labelmodel/run_labelmodel.py ran Apache-2.0 (permissive) · 9f7ae362c69b5fbf · report fixture: returned tuple · level recorded 2026-09-24
mse in 78 papers
census: 5 implementations from 6 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
taglets/labelmodel/amcl_helper.py ran fingerprinted Apache-2.0 (permissive) · 4b28a291d70a1b4b · report fixture: signature y_1:ndarray/1/float, y_2:ndarray/1/float · returned float64 · level recorded 2026-09-24
tf2th in 19 papers taglets/models/resnetv2.py ran · violated contract fingerprinted Apache-2.0 (permissive) · 7023eb86801bb3a3 · report fixture: signature conv_weights:ndarray/4/float · returned Tensor · level recorded 2026-09-22
get_data in 223 papers taglets/labelmodel/run_labelmodel.py unverified Apache-2.0 (permissive) · 25fc1400e3579d6f · report

Tasks

Few-Shot LearningFine-Grained Image ClassificationImage ClassificationOut-of-Distribution GeneralizationRepresentation Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Fine-Grained Image Classification Oxford 102 Flowers BiT-L (ResNet) Accuracy 99.63% #2 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers BiT-L (ResNet) Top-1 Error Rate 0.37 #2 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers BiT-M (ResNet) Accuracy 99.30% #4 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford 102 Flowers BiT-M (ResNet) Top-1 Error Rate 0.70 #4 of 25 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pets BiT-L (ResNet) Accuracy 96.62 #2 of 19 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pets BiT-L (ResNet) Top-1 Error Rate 3.38% #2 of 19 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pets BiT-M (ResNet) Accuracy 94.47 #5 of 19 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pets BiT-M (ResNet) Top-1 Error Rate 5.53% #5 of 19 Archive leaderboard report
Image Classification CIFAR-10 BiT-L (ResNet) Percentage correct 99.37 #7 of 265 Archive leaderboard report
Image Classification CIFAR-10 BiT-M (ResNet) Percentage correct 98.91 #26 of 265 Archive leaderboard report
Image Classification CIFAR-100 BiT-L (ResNet) Percentage correct 93.51 #8 of 211 Archive leaderboard report
Image Classification CIFAR-100 BiT-M (ResNet) Percentage correct 92.17 #15 of 211 Archive leaderboard report
Image Classification Flowers-102 BiT-L (ResNet) Accuracy 99.63 #8 of 52 Archive leaderboard report
Image Classification Flowers-102 BiT-M (ResNet) Accuracy 99.30 #12 of 52 Archive leaderboard report
Image Classification ImageNet BiT-L (ResNet) Top 1 Accuracy 87.54% #80 of 1060 Archive leaderboard report
Image Classification ImageNet BiT-L (ResNet) Top 5 Accuracy 98.46 #80 of 1060 Archive leaderboard report
Image Classification ImageNet BiT-M (ResNet) Number of params 928M #237 of 1060 Archive leaderboard report
Image Classification ImageNet BiT-M (ResNet) Top 1 Accuracy 85.39% #237 of 1060 Archive leaderboard report
Image Classification ImageNet ReaL BiT-L Accuracy 90.54% #17 of 57 Archive leaderboard report
Image Classification ImageNet ReaL BiT-L Params 928M #17 of 57 Archive leaderboard report
Image Classification ImageNet ReaL BiT-M Accuracy 89.02% #25 of 57 Archive leaderboard report
Image Classification ObjectNet BiT-L (ResNet-152x4) Top-1 Accuracy 58.7 #21 of 106 Archive leaderboard report
Image Classification ObjectNet BiT-L (ResNet-152x4) Top-5 Accuracy 80 #21 of 106 Archive leaderboard report
Image Classification ObjectNet BiT-M (ResNet-152x4) Top-1 Accuracy 47.0 #33 of 106 Archive leaderboard report
Image Classification ObjectNet BiT-M (ResNet-152x4) Top-5 Accuracy 69 #33 of 106 Archive leaderboard report
Image Classification ObjectNet BiT-S (ResNet-152x4) Top-1 Accuracy 36.0 #50 of 106 Archive leaderboard report
Image Classification ObjectNet BiT-S (ResNet-152x4) Top-5 Accuracy 57 #50 of 106 Archive leaderboard report
Image Classification ObjectNet (Bounding Box) BiT-L (ResNet) Top 5 Accuracy 85.1 #1 of 4 Archive leaderboard report
Image Classification ObjectNet (Bounding Box) BiT-M (ResNet) Top 5 Accuracy 76.0 #2 of 4 Archive leaderboard report
Image Classification ObjectNet (Bounding Box) BiT-S (ResNet) Top 5 Accuracy 64.4 #3 of 4 Archive leaderboard report
Image Classification OmniBenchmark BiT-M Average Top-1 Accuracy 40.4 #7 of 22 Archive leaderboard report
Image Classification VTAB-1k BiT-L (50 hypers/task) Top-1 Accuracy 78.72 #2 of 34 Archive leaderboard report
Image Classification VTAB-1k BiT-L Top-1 Accuracy 76.3 #6 of 34 Archive leaderboard report
Image Classification VTAB-1k BiT-M Top-1 Accuracy 70.6 #11 of 34 Archive leaderboard report
Image Classification VTAB-1k BiT-S Top-1 Accuracy 66.9 #15 of 34 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

1x1 ConvolutionAverage PoolingBottleneck Residual BlockConvolutionFixResGlobal Average PoolingGroup NormalizationKaiming InitializationMax PoolingMixupRandom Horizontal FlipRandom Resized CropReLUResidual BlockResidual ConnectionSGD with MomentumWeight Standardization

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