Browse › Self-Supervised Image Classification › ImageNet (finetuned)

Self-Supervised Image Classification archive 2025-07-28

ImageNet (finetuned) Benchmark (Self-Supervised Image Classification)

65 rows 64 with code listed 2 metrics Dataset page

Syntology Since the archive (Syntology): none placed; 3 newer papers judged to report on this board so far.

The archive's category list for this table covers 11 areas (Adversarial, Audio, Computer Code, Computer Vision, Medical, Methodology, Miscellaneous, Music, Natural Language Processing, Reasoning, Speech), which is its placeholder rather than an assignment; no area is shown in the breadcrumb. archive 2025-07-28

This is the task of image classification using representations learnt with self-supervised learning. Self-supervised methods generally involve a pretext task that is solved to learn a good representation and a loss function to learn with. One example of a loss function is an autoencoder based loss where the goal is reconstruction of an image pixel-by-pixel. A more popular recent example is a contrastive loss, which measure the similarity of sample pairs in a representation space, and where there can be a varying target instead of a fixed target to reconstruct (as in the case of autoencoders).

A common evaluation protocol is to train a linear classifier on top of (frozen) representations learnt by self-supervised methods. The leaderboards for the linear evaluation protocol can be found below. In practice, it is more common to fine-tune features on a downstream task. An alternative evaluation protocol therefore uses semi-supervised learning and finetunes on a % of the labels. The leaderboards for the finetuning protocol can be accessed here.

You may want to read some blog posts before reading the papers and checking the leaderboards:

There is also Yann LeCun's talk at AAAI-20 which you can watch here (35:00+).

The archive carries no text for this table; the description above is the archive's text for the task Self-Supervised Image Classification. archive 2025-07-28

Over time archive 2025-07-28

Direction inferred from the metric name, not from the archive: Top 1 Accuracy (higher is better), Number of Params (lower is better). Points are placed at the row's paper date, not an evaluation date; 65 of 65 rows carry one.

The chart needs JavaScript; the table below carries every value.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 DINOv2 (ViT-g/14, 448) 88.9%1100M ✓ Paper Code 2023 community repositories only · 39 ran (of which 13 constructed an object rather than computing a result; 32 with no instrument failure: 2 honoured, 0 violated, 30 with no contract checked; 7 where Syntology's instrument failed) · 7 unverified report
2 PercMAE (ViT-L, dVAE) 88.6%307M ✓ Paper Code 2022 linked, not harvested report
3 DINOv2 (ViT-g/14) 88.5%1100M ✓ Paper Code 2023 community repositories only · 39 ran (of which 13 constructed an object rather than computing a result; 32 with no instrument failure: 2 honoured, 0 violated, 30 with no contract checked; 7 where Syntology's instrument failed) · 7 unverified report
4 PeCo(ViT-H/14, 448) 88.3%632M – Paper Code 2021 linked, not harvested report
5 PercMAE (ViT-L) 88.1%307M – Paper Code 2022 linked, not harvested report
6 dBOT (ViT-H/14) 88.0%632M – Paper Code 2022 official (archive's flag): 11 ran · 11 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 1 violated, 7 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified report
7 MAE (ViT-H/14, 448) 87.8%632M – Paper Code 2021 official: no sample here; runs from other or unrecorded repositories · 86 ran (of which 40 constructed an object rather than computing a result; 68 with no instrument failure: 8 honoured, 0 violated, 60 with no contract checked; 18 where Syntology's instrument failed) · 51 unverified report
8 iBOT(ViT-L/16, 512) 87.8%307M ✓ Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
9 MAE + AugSub finetune (ViT-H/14) 87.2%632M – Paper Code 2023 linked, not harvested report
10 SimMIM (SwinV2-H, 512) 87.1%658M – Paper Code 2021 official (archive's flag): 1 ran · 9 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified report
11 MAE (ViT-H/14) 86.9% – Paper Code 2021 official: no sample here; runs from other or unrecorded repositories · 86 ran (of which 40 constructed an object rather than computing a result; 68 with no instrument failure: 8 honoured, 0 violated, 60 with no contract checked; 18 where Syntology's instrument failed) · 51 unverified report
12 iBOT(ViT-L/16) 86.6%307M ✓ Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
13 TEC_MAE (ViT-L/16, 224) 86.5% – Paper Code 2022 official (archive's flag): 6 ran · 6 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 6 samples that ran constructed an object rather than computing a result report
14 BEiT-L (ViT) 86.3%307M ✓ Paper Code 2021 official (archive's flag): 1 ran · 6 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified report
15 CAE (ViT-L/16) 86.3%307M – Paper Code 2022 linked, not harvested report
16 MIRL (ViT-B-48) 86.2%341M – Paper Code 2023 linked, not harvested report
17 MAE + AugSub finetune (ViT-L/16) 86.1%304M – Paper Code 2023 linked, not harvested report
18 SparK (ConvNeXt-Large, 384) 86.0%198M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
19 BootMAE(ViT-L) 85.9%307M – Paper Code 2022 official (archive's flag): 6 ran · 6 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; every one of the 6 samples that ran constructed an object rather than computing a result report
20 SEER (Regnet10B) 85.8%10000M ✓ Paper Code 2022 linked, not harvested report
21 MaskFeat (ViT-L) 85.7%307M – Paper Code 2021 linked, not harvested report
22 OFA (Large) 85.6%473M – Paper Code 2022 official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
23 SparK (ConvNeXt-Large) 85.4%198M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
24 SimMIM (Swin-L) 85.4%197M – Paper Code 2021 official (archive's flag): 1 ran · 9 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified report
25 Mugs (ViT-L/16) 85.2%307M – Paper Code 2022 official (archive's flag): 19 ran · 19 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 1 honoured, 1 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified report
26 iBOT (ViT-L/16) 84.8%307M – Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
27 MIRL (ViT-S-54) 84.8%96M – Paper Code 2023 linked, not harvested report
28 ConvNeXt-Base (SparK pre-training) 84.8%89M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
29 BEiT-B (ViT) 84.6%86M ✓ Paper Code 2021 official (archive's flag): 1 ran · 6 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified report
30 A2MIM+ (ViT-B) 84.5% – Paper Code 2022 linked, not harvested report
31 iBOT (ViT-B/16) 84.4%85M – Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
32 Mugs (ViT-B/16) 84.3%85M – Paper Code 2022 official (archive's flag): 19 ran · 19 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 1 honoured, 1 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified report
33 SEER (RegNetY-256GF) 84.2%1.3B ✓ Paper Code 2021 linked, not harvested report
34 A2MIM (ViT-B) 84.2% – Paper Code 2022 linked, not harvested report
35 MoCo v3 (ViT-L/16) 84.1%304M – Paper Code 2021 official: no sample here; runs from other or unrecorded repositories · 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 report
36 mc-BEiT (ViT-B/16) 84.1%86M – Paper Code 2022 official (archive's flag): 3 ran · 3 ran (of which 3 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) · 2 unverified; every one of the 3 samples that ran constructed an object rather than computing a result report
37 ConvNeXt-Small (SparK pre-training) 84.1%50M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
38 SimMIM (Swin-B) 84.0%88M – Paper Code 2021 official (archive's flag): 1 ran · 9 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified report
39 iBOT (ViT-B/16) 84.0%85M – Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
40 EsViT (Swin-B) 83.9%87M – Paper Code 2021 official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified report
41 MAE + AugSub finetune (ViT-B/16) 83.9%87M – Paper Code 2023 linked, not harvested report
42 SEER (RegNetY-128GF) 83.8%693M ✓ Paper Code 2021 linked, not harvested report
43 SimMIM (ViT-B/16) 83.8%85M – Paper Code 2021 official (archive's flag): 1 ran · 9 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified report
44 MoCo v3 (ViT-B/16) 83.2%86M – Paper Code 2021 official: no sample here; runs from other or unrecorded repositories · 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 report
45 DAMA (ViT-B/16) 83.2% – Paper Code 2022 linked, not harvested report
46 SimCLRv2 (ResNet-152, 3×+SK) 83.1%795M – Paper Code 2020 community repositories only · 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified report
47 ResNet-200 (SparK pre-training) 83.1%65M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
48 DINO (ViT-B/16) 82.8%85M – Paper Code 2021 official: no sample here; runs from other or unrecorded repositories · 14 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 1 violated, 9 with no contract checked; 4 where Syntology's instrument failed) · 6 unverified report
49 ResNet-152 (SparK pre-training) 82.7%60M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
50 Mugs (ViT-S/16) 82.6%21M – Paper Code 2022 official (archive's flag): 19 ran · 19 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 1 honoured, 1 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified report
51 A2MIM+ (ViT-S) 82.4% – Paper Code 2022 linked, not harvested report
52 ResNet-101 (SparK pre-training) 82.2%44M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
53 A2MIM (ViT-S) 82.2% – Paper Code 2022 linked, not harvested report
54 SwAV (ResNeXt-101-32x16d) 82.0%193M ✓ Paper Code 2020 community repositories only · 13 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 1 violated, 4 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified report
55 ResNet-50 (SparK pre-training) 80.6%26M – Paper Code 2023 official (archive's flag): 9 ran · 13 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified report
56 A2MIM+ (ResNet-50 RSB-A2) 80.5% – Paper Code 2022 linked, not harvested report
57 A2MIM (ResNet-50 RSB-A2) 80.4% – Paper Code 2022 linked, not harvested report
58 A2MIM+ (ResNet-50 RSB-A3) 78.9% – Paper Code 2022 linked, not harvested report
59 A2MIM (ResNet-50 RSB-A3) 78.8% – Paper Code 2022 linked, not harvested report
60 DnC (Resnet-50) 78.2% – Paper – 2021 no code linked report
61 SwAV (Resnet-50) 77.8%182M ✓ Paper Code 2020 community repositories only · 13 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 1 violated, 4 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified report
62 MoCo (Resnet-50) 77.3% ✓ Paper Code 2019 community repositories only · 28 ran (of which 16 constructed an object rather than computing a result; 23 with no instrument failure: 0 honoured, 0 violated, 23 with no contract checked; 5 where Syntology's instrument failed) · 14 unverified report
63 SimCLR (Resnet-50) 77.2% ✓ Paper Code 2020 official (archive's flag): 2 ran · 115 ran (of which 33 constructed an object rather than computing a result; 89 with no instrument failure: 1 honoured, 1 violated, 87 with no contract checked; 26 where Syntology's instrument failed) · 22 unverified report
64 MoCo (Resnet-50) 77.0% – Paper Code 2019 community repositories only · 28 ran (of which 16 constructed an object rather than computing a result; 23 with no instrument failure: 0 honoured, 0 violated, 23 with no contract checked; 5 where Syntology's instrument failed) · 14 unverified report
65 DeeperCluster (VGG16) 74.9%138M ✓ Paper Code 2019 linked, not harvested report

All 65 rows shown. 65 link to a paper page on this site; 15 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”: of the code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and U are unverified (harvested, no recorded run). Of the N, C constructed an object rather than computing a result (a part of N, never taken away from it), K ran with no instrument failure (honoured or violated the contract Syntology drafted, or ran with no contract checked) and I are failures of Syntology's instrument, not of the code (a wrong contract draft, or a contract check that could not run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 41 rows have a graph line, from 19 distinct papers. Rows whose paper has a sample that ran: 36 (34 with a run with no instrument failure, 2 where every run was a failure of Syntology's instrument). Papers: 18 with a sample that ran: 17 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument. Whose code ran. When the archive marks a repository official for the row's paper, the cell starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code): “official (archive's flag): N ran” counts the samples from it that ran, instrument failures included; “official: harvested, nothing ran”; when it has no sample for this paper and other samples ran, “community repositories only” if every one came from a repository the archive lists for the paper but neither marks official nor records as mentioned in it, or one Syntology harvested outside the archive's links, and “official: no sample here; runs from other or unrecorded repositories” if any came from a repository named in the paper or found in its text by Syntology, or has no repository recorded; “official: harvested for another paper” (no sample from it is linked to this paper) or “official: not harvested”. No state is printed when the archive marks none official. The cell's tooltip names the repositories the samples that ran came from. Of the 41 rows with a graph line, 41 have an archive-official repository: on 24 a sample from it ran (23 with a run with no instrument failure), on 7 the samples that ran came from community repositories only, and on 5 from other or unrecorded repositories. The filters above hide rows and never renumber them; the first two count, by default, only a run with no instrument failure, and offer any run, instrument failures included. Counting each paper once, of 508 samples: 377 ran (of which 130 constructed an object rather than computing a result; 299 with no instrument failure: 16 honoured, 5 violated, 278 with no contract checked; 78 where Syntology's instrument failed) · 131 unverified. Separately, 200 of those 508 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. Per-sample status is on the paper page; each Ran cell links to that paper's sample list. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Since the archive Syntology

Of the 8,888 newer papers whose tables Syntology has read (8,886 checked in full and 2 with a leaderboard still awaiting a verdict), 54 had this board among the leaderboards the extractor considered for their tables (it keeps at most 10 boards per paper and 4 per dataset; the lists of 1,550 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and 3 proposed a result; of those 3, 3 gave no cell that passed the extractor's own checks, 0 were rejected by the independent check, 0 await a verdict and 0 were placed. Syntology has checked 8,886 of the 18,637 papers on this site that are newer than the archive (for 3,717 of them no archive leaderboard matched the paper's tables, so there was nothing further to check); 733 were read and have nothing to place (arXiv has no HTML version of the paper, or that version has no tables), 41 could not be read (the extractor's reply could not be parsed), and results from the other 8,977 appear after they are checked.

Newer papers whose tables Claude Sonnet 4.5 judged to report on this benchmark (a model's judgement, not a result; newest first by arXiv id)

Paper arXiv date Ran Syntology Outcome
Stylistic-STORM (ST-STORM) : Perceiving the Semantic Nature of Appearance arXiv:2604.16086 17 Apr 2026 13 harvested: 12 ran (of which 4 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 7 where Syntology's instrument failed) · 1 unverified refused by a rule: the value is not on the scale of the leaderboard column's archive values
Soft Equivariance Regularization for Invariant Self-Supervised Learning arXiv:2603.06693 4 Mar 2026 1 harvested: 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) · 0 unverified refused by a rule: the table's column header does not name the leaderboard's metric
ToaSt: Token Channel Selection and Structured Pruning for Efficient ViT arXiv:2602.15720 17 Feb 2026 6 harvested: 3 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; 3 where Syntology's instrument failed) · 3 unverified refused by a rule: the table's column header does not name the leaderboard's metric

Syntology 3 newer papers judged to report on this board: the papers counted above as having proposed a result (0 placed); every one is listed here and in the JSON twin. The order is the arXiv identifier's, newest first, not a rank. The judgement is the model's, read from the paper's own tables; the outcome is what Syntology's checks did with it, and no value from a paper that was not placed is printed. “refused by a rule” names the extractor's rule; “no cell passed the extractor's checks” means the extractor's record holds no cell for this board, passed or refused. The 3 counted above as giving no cell that passed the extractor's own checks are the 3 marked “refused by a rule”. Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

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