Papers › An Empirical Study of Training Self-Supervised Vision Transformers

An Empirical Study of Training Self-Supervised Vision Transformers

5 Apr 2021ICCV 2021 10arXiv:2104.02057archive 2025-07-28

Xinlei Chen, Saining Xie, Kaiming He

This paper does not describe a novel method. Instead, it studies a straightforward, incremental, yet must-know baseline given the recent progress in computer vision: self-supervised learning for Vision Transformers (ViT). While the training recipes for standard convolutional networks have been highly mature and robust, the recipes for ViT are yet to be built, especially in the self-supervised scenarios where training becomes more challenging. In this work, we go back to basics and investigate the effects of several fundamental components for training self-supervised ViT. We observe that instability is a major issue that degrades accuracy, and it can be hidden by apparently good results. We reveal that these results are indeed partial failure, and they can be improved when training is made more stable. We benchmark ViT results in MoCo v3 and several other self-supervised frameworks, with ablations in various aspects. We discuss the currently positive evidence as well as challenges and open questions. We hope that this work will provide useful data points and experience for future research.

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

The archive's official repository (facebookresearch/moco-v3) was not harvested; the 1 sample below comes from 1 community repository, and 1 more is identical to code Syntology harvested from another paper's repository.

Syntology 2 code samples: 1 from 1 repository linked to this paper, 1 identical to code harvested from another paper's repository. Across all 2: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code. Every run on this page is a failure of Syntology's instrument, not of the code.

By repository: community (archive-listed): 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. 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 4 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 2 samples that ran has no recorded date.

facebookresearch/moco-v3 officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
CupidJay/MoCov3-pytorch mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
KAIST-AILab/MaskedAutoencoder-Jax mentioned on GitHubjaxApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Multimodal Masked Autoencoders Learn Transferable Representations (arXiv:2205.14204; that paper's own run record: 12 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
Westlake-AI/openmixup mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to mixup: Beyond Empirical Risk Minimization (arXiv:1710.09412; that paper's own run record: 36 ran (of which 5 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 22 where Syntology's instrument failed) · 11 unverified), Visualizing the Loss Landscape of Neural Nets (arXiv:1712.09913; that paper's own run record: 3 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified), CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features (arXiv:1905.04899; that paper's own run record: 17 ran (of which 6 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 7 where Syntology's instrument failed) · 7 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
oliverrensu/sdmp mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to A Simple Data Mixing Prior for Improving Self-Supervised Learning (arXiv:2206.07692; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
oneflow-inc/libai mentioned on GitHubApache-2.0Syntology: not harvested report
open-mmlab/mmselfsup mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to PixMIM: Rethinking Pixel Reconstruction in Masked Image Modeling (arXiv:2303.02416; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
xiyue-wang/transpath mentioned on GitHubpytorchGPL-3.0Syntology: not harvested report
toliz/moco-v3 pytorchSyntology: 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

2 samples harvested: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Every run on this page is a failure 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 · our draft was wrong
1ran; the contract check could not run

Licence: 2 of the 2 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 CupidJay/MoCov3-pytorch. 1 sample is listed last, in a group without a repository. “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.

CupidJay/MoCov3-pytorch community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
ctr main_moco_pretraining_v3.py ran · our draft was wrong no licence file found · pointer only · c73266677f2758b5 · report fixture: returned Tensor · level recorded 2026-09-22
Repository not located 1 sample, 1 ran (1 instrument-failed) · 0 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
repository not located ran; the contract check could not run (RAISES) · metamorphic tier: invariant licence of this copy not recorded · 983576fa67f4f01e · report fixture: signature output:Tensor/2/float, target:Tensor/1/int, topk:tuple/1/int · returned list · level recorded, date not recorded

Tasks

Out-of-Distribution GeneralizationSelf-Supervised Image ClassificationSelf-Supervised Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-BN-L/7) Number of Params 304M #18 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-BN-L/7) Top 1 Accuracy 81.0% #18 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-BN-H) Number of Params 700M #37 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-BN-H) Top 1 Accuracy 79.1% #37 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-H) Number of Params 632M #47 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-H) Top 1 Accuracy 78.1% #47 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-L) Number of Params 307M #51 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-L) Top 1 Accuracy 77.6% #51 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-B/16) Number of Params 86M #58 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet MoCo v3 (ViT-B/16) Top 1 Accuracy 76.7% #58 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) MoCo v3 (ViT-L/16) Number of Params 304M #35 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) MoCo v3 (ViT-L/16) Top 1 Accuracy 84.1% #35 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) MoCo v3 (ViT-B/16) Number of Params 86M #44 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) MoCo v3 (ViT-B/16) Top 1 Accuracy 83.2% #44 of 65 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

Introduced by this paper: MoCo v3

Batch NormalizationInfoNCEMoCoMoCo v3

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