Browse State-of-the-Art › Learning with noisy labels

Learning with noisy labels

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

AdversarialComputer VisionMedicalNatural Language Processing

Learning with noisy labels means When we say "noisy labels," we mean that an adversary has intentionally messed up the labels, which would have come from a "clean" distribution otherwise. This setting can also be used to cast learning from only positive and unlabeled data.

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.

Syntology 2 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): ANIMAL (1 judged, 0 placed), Clothing1M (1 judged, 0 placed), Food-101 (1 judged, 0 placed), CIFAR-10N (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
CIFAR-10N-Aggregate (26 rows) ProMix ProMix: Combating Label Noise via Maximizing Clean Sample Utility code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
CIFAR-10N-Worst (25 rows) ProMix ProMix: Combating Label Noise via Maximizing Clean Sample Utility code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
CIFAR-100N (24 rows) PGDF Sample Prior Guided Robust Model Learning to Suppress Noisy Labels code — Compare
CIFAR-10N-Random1 (24 rows) ProMix ProMix: Combating Label Noise via Maximizing Clean Sample Utility code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
CIFAR-10N-Random2 (23 rows) PSSCL PSSCL: A progressive sample selection framework with contrastive... code — Compare
CIFAR-10N-Random3 (23 rows) PSSCL PSSCL: A progressive sample selection framework with contrastive... code — Compare
ANIMAL (19 rows) Jigsaw-ViT Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer code — Compare
Clothing1M (5 rows) Knockoffs-SPR Knockoffs-SPR: Clean Sample Selection in Learning with Noisy Labels code — Compare
Red MiniImageNet 20% label noise (4 rows) NCR (ResNet-18) Learning with Neighbor Consistency for Noisy Labels code Syntology official (archive's flag): 4 ran · 4 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; 4 where Syntology's instrument failed) · 0 unverified Compare
Red MiniImageNet 40% label noise (4 rows) NCR (ResNet-18) Learning with Neighbor Consistency for Noisy Labels code Syntology official (archive's flag): 4 ran · 4 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; 4 where Syntology's instrument failed) · 0 unverified Compare
Red MiniImageNet 80% label noise (4 rows) NCR (ResNet-18) Learning with Neighbor Consistency for Noisy Labels code Syntology official (archive's flag): 4 ran · 4 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; 4 where Syntology's instrument failed) · 0 unverified Compare
Food-101 (3 rows) LongReMix LongReMix: Robust Learning with High Confidence Samples in a Noisy... code — Compare
Red MiniImageNet 60% label noise (3 rows) InstanceGM-SS Instance-Dependent Noisy Label Learning via Graphical Modelling code — Compare
Chaoyang (1 row) HSANR Hard Sample Aware Noise Robust Learning for Histopathology Image... code — Compare
CIFAR-10 (1 row) InstanceGM Instance-Dependent Noisy Label Learning via Graphical Modelling code — Compare
CIFAR-100 (1 row) InstanceGM Instance-Dependent Noisy Label Learning via Graphical Modelling code — Compare
CIFAR-10N (1 row) ProMix ProMix: Combating Label Noise via Maximizing Clean Sample Utility code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Clothing1M (using clean data) (1 row) ResNet50 UNICON: Combating Label Noise Through Uniform Selection and... code Syntology official (archive's flag): 12 ran · 12 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified Compare
COCO-WAN (1 row) Mask R-CNN (ResNet-50-FPN) Benchmarking Label Noise in Instance Segmentation: Spatial Noise Matters code Syntology official (archive's flag): 2 ran · 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 Compare
mini WebVision 1.0 (1 row) ILL Imprecise Label Learning: A Unified Framework for Learning with... code — 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

No subtask under this task in the archive's task tree.

Parent tasks archive 2025-07-28

Papers archive 2025-07-28

archive papers tagged: 249 · with a code link: 143 · where Syntology ran a sample: 50 (41 with a run with no instrument failure, 9 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (50 of 249 tagged: 41 with a run with no instrument failure, 9 where every run was a failure of Syntology's instrument)

Page 1 of 3: papers 1 to 100 of 249, 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 41 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