Browse State-of-the-Art › Long-tail Learning

Long-tail Learning

93 papers with code · 20 benchmarks · 16 datasets archive 2025-07-28

Methodology

Long-tailed learning, one of the most challenging problems in visual recognition, aims to train well-performing models from a large number of images that follow a long-tailed class distribution.

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

20 leaderboard tables shown for this task, 20 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 20 until expanded.

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
ImageNet-LT (69 rows) LIFT (ViT-L/14) Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts code Syntology official (archive's flag): 5 ran · 5 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; 4 where Syntology's instrument failed) · 3 unverified Compare
CIFAR-100-LT (ρ=100) (66 rows) LPT LPT: Long-tailed Prompt Tuning for Image Classification code Syntology official (archive's flag): 2 ran · 2 ran (of which 2 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) · 0 unverified; every one of the 2 samples that ran constructed an object rather than computing a result Compare
CIFAR-10-LT (ρ=10) (50 rows) GLMC+MaxNorm (ResNet-34, channel x4) Global and Local Mixture Consistency Cumulative Learning for... code Syntology official (archive's flag): 2 ran · 7 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; 4 where Syntology's instrument failed) · 3 unverified Compare
iNaturalist 2018 (43 rows) LIFT (ViT-L/14@336px) Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts code Syntology official (archive's flag): 5 ran · 5 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; 4 where Syntology's instrument failed) · 3 unverified Compare
CIFAR-100-LT (ρ=10) (31 rows) LIFT (ViT-B/16, ImageNet-21K pre-training) Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts code Syntology official (archive's flag): 5 ran · 5 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; 4 where Syntology's instrument failed) · 3 unverified Compare
Places-LT (29 rows) LIFT (ViT-L/14) Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts code Syntology official (archive's flag): 5 ran · 5 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; 4 where Syntology's instrument failed) · 3 unverified Compare
CIFAR-10-LT (ρ=100) (28 rows) GLMC+MaxNorm (ResNet-34, channel x4) Global and Local Mixture Consistency Cumulative Learning for... code Syntology official (archive's flag): 2 ran · 7 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; 4 where Syntology's instrument failed) · 3 unverified Compare
CIFAR-100-LT (ρ=50) (25 rows) LIFT (ViT-B/16, ImageNet-21K pre-training) Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts code Syntology official (archive's flag): 5 ran · 5 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; 4 where Syntology's instrument failed) · 3 unverified Compare
MIMIC-CXR-LT (15 rows) Decoupling (cRT) Long-Tailed Classification of Thorax Diseases on Chest X-Ray: A... code — Compare
NIH-CXR-LT (15 rows) Decoupling (cRT) Long-Tailed Classification of Thorax Diseases on Chest X-Ray: A... code — Compare
COCO-MLT (13 rows) LMPT(ViT-B/16) LMPT: Prompt Tuning with Class-Specific Embedding Loss for... code Syntology official (archive's flag): 4 ran · 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) · 5 unverified Compare
VOC-MLT (13 rows) LMPT(ViT-B/16) LMPT: Prompt Tuning with Class-Specific Embedding Loss for... code Syntology official (archive's flag): 4 ran · 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) · 5 unverified Compare
CIFAR-10-LT (ρ=50) (8 rows) GLMC + SAM Escaping Saddle Points for Effective Generalization on... code — Compare
ImageNet-GLT (6 rows) RIDE + IFL Invariant Feature Learning for Generalized Long-Tailed Classification code Syntology official (archive's flag): 2 ran · 2 ran (of which 2 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) · 0 unverified; every one of the 2 samples that ran constructed an object rather than computing a result Compare
EGTEA (3 rows) CDB-loss (3D- ResNeXt101) Class-Wise Difficulty-Balanced Loss for Solving Class-Imbalance code Syntology official: harvested, nothing ran · 0 ran · 2 unverified Compare
CIFAR-10-LT (ρ=200) (2 rows) LDAM + DRW + SAM Escaping Saddle Points for Effective Generalization on... code — Compare
CIFAR-100-LT (ρ=200) (2 rows) PaCo + SAM Escaping Saddle Points for Effective Generalization on... code — Compare
CelebA-5 (1 row) OPeN (WideResNet-28-10) Pure Noise to the Rescue of Insufficient Data: Improving... code — Compare
Lot-insts (1 row) Character-BERT+RS Text Classification in the Wild: a Large-scale Long-tailed Name... code — Compare
mini-ImageNet-LT (1 row) TailCalibX Feature Generation for Long-tail Classification code Syntology official (archive's flag): 5 ran · 5 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; 1 where Syntology's instrument failed) · 1 unverified Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line 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”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

16 datasets whose archive record lists this task, ordered by the archive's paper count.

Subtasks archive 2025-07-28

1 subtask in the archive's task tree.

Parent tasks archive 2025-07-28

Papers archive 2025-07-28

archive papers tagged: 131 · with a code link: 93 · where Syntology ran a sample: 50 (44 with a run with no instrument failure, 6 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (50 of 131 tagged: 44 with a run with no instrument failure, 6 where every run was a failure of Syntology's instrument)

Page 1 of 2: papers 1 to 100 of 131, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line 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”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 52 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

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