Papers › Open-set Label Noise Can Improve Robustness Against Inherent Label Noise

Open-set Label Noise Can Improve Robustness Against Inherent Label Noise

21 Jun 2021NeurIPS 2021 12arXiv:2106.10891archive 2025-07-28

Hongxin Wei, Lue Tao, Renchunzi Xie, Bo An

Learning with noisy labels is a practically challenging problem in weakly supervised learning. In the existing literature, open-set noises are always considered to be poisonous for generalization, similar to closed-set noises. In this paper, we empirically show that open-set noisy labels can be non-toxic and even benefit the robustness against inherent noisy labels. Inspired by the observations, we propose a simple yet effective regularization by introducing Open-set samples with Dynamic Noisy Labels (ODNL) into training. With ODNL, the extra capacity of the neural network can be largely consumed in a way that does not interfere with learning patterns from clean data. Through the lens of SGD noise, we show that the noises induced by our method are random-direction, conflict-free and biased, which may help the model converge to a flat minimum with superior stability and enforce the model to produce conservative predictions on Out-of-Distribution instances. Extensive experimental results on benchmark datasets with various types of noisy labels demonstrate that the proposed method not only enhances the performance of many existing robust algorithms but also achieves significant improvement on Out-of-Distribution detection tasks even in the label noise setting.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

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

Syntology 29 code samples: 28 from 3 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 29: 17 ran (of which 5 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 7 where Syntology's instrument failed) · 12 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 13 samples from 1 repository, 9 ran (7 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 15 samples from 2 repositories, 7 ran (3 with no instrument failure; 4 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 1 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.

hongxin001/ODNL officialmentioned in papermentioned on GitHubpytorchSyntology: 13 samples, 9 ran (2 instrument-failed) · 4 unverified report
hongxin001/JoCoR mentioned on GitHubpytorchSyntology: 6 samples, 5 ran (3 instrument-failed) · 1 unverified report
hongxin001/logitnorm_ood mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Mitigating Neural Network Overconfidence with Logit Normalization (arXiv:2205.09310; that paper's own run record: 1 ran (of which 1 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; the one sample that ran constructed an object rather than computing a result); nothing checks that they implement this paper's method. report
hongxin001/open-sampling mentioned on GitHubpytorchSyntology: 9 samples, 2 ran (1 instrument-failed) · 7 unverified 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

29 samples harvested: 17 ran (of which 5 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 7 where Syntology's instrument failed) · 12 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 · honoured contract
1ran · violated contract
6ran · our draft was wrong
1ran; the contract check could not run
8ran
12unverified

Licence: 29 of the 29 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 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 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”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

hongxin001/ODNL official repository 13 samples, 9 ran (2 instrument-failed) · 4 unverified
BasicBlock in 126 papers
census: 6 implementations from 6 papers, 5 distinct outputs on one shared input
algorithms/odnl.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 68c820cf9fac1d45 · report fixture: returned BasicBlock · level recorded 2026-09-22
DependentLabelGenerator algorithms/odnl.py ran no licence file found · pointer only · 3aa8c19f6a93cc4f · report fixture: returned DependentLabelGenerator · level recorded 2026-09-22
SVHN algorithms/odnl.py ran no licence file found · pointer only · a48cbe21ee4b47c7 · report fixture: returned SVHN · level recorded 2026-09-22
SingleModel algorithms/odnl.py ran no licence file found · pointer only · 5ef095faf1a178df · report fixture: returned SingleModel · level recorded 2026-09-22
WideResNet in 13 papers algorithms/odnl.py ran · metamorphic tier: deterministic no licence file found · pointer only · bbf25f185bcee0a2 · report fixture: returned WideResNet · level recorded 2026-09-22
build_model in 469 papers algorithms/odnl.py ran; the contract check could not run (DEP_MISSING) · metamorphic tier: deterministic no licence file found · pointer only · 00092d7deaae3c5a · report fixture: signature model_type:str/0/str, num_classes:int/0/int, device:device/0/?, args:Args/0/? · returned WideResNet · level recorded 2026-09-22
build_ood_noise algorithms/odnl.py ran · our draft was wrong no licence file found · pointer only · 6463c81eb9a79759 · report fixture: signature noise_type:str/0/str, ood_num_examples:int/0/int, num_to_avg:int/0/int · returned TensorDataset · level recorded 2026-09-22
check_integrity in 37 papers algorithms/odnl.py ran · violated contract fingerprinted no licence file found · pointer only · 379a713ff6573163 · report fixture: signature fpath:str/0/str, md5:str/0/str · returned bool · level recorded 2026-09-22
get_dataset_normlize algorithms/odnl.py ran · honoured contract no licence file found · pointer only · 331a7108385c91ba · report fixture: signature dataset:str/0/str · returned tuple · level recorded 2026-09-22
CIFAR100 in 7 papers algorithms/odnl.py unverified no licence file found · pointer only · b83e460dc32b7006 · report
ODNL algorithms/odnl.py unverified no licence file found · pointer only · b998044858807188 · report
build_dataset in 151 papers algorithms/odnl.py unverified no licence file found · pointer only · b3a727f63915052f · report
download_url in 32 papers algorithms/odnl.py unverified no licence file found · pointer only · 0105f183fe8b9666 · report
hongxin001/JoCoR community (archive-listed) 6 samples, 5 ran (3 instrument-failed) · 1 unverified
CNN in 23 papers algorithm/jocor.py ran · metamorphic tier: invariant no licence file found · pointer only · c15655fa5cf01c52 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
MLPNet algorithm/jocor.py ran · metamorphic tier: invariant no licence file found · pointer only · dca203cfebebf4d7 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
accuracy in 872 papers
census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket
algorithm/jocor.py ran · our draft was wrong no licence file found · pointer only · 5654f848f08debcf · report fixture: signature logit:Tensor/2/float, target:Tensor/1/int, topk:tuple/1/int · returned Tensor · level recorded 2026-09-22
kl_loss_compute algorithm/jocor.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 75feb4c4aa30068a · report fixture: signature pred:Tensor/2/float, soft_targets:Tensor/2/float, reduce:bool/0/bool · returned Tensor · level recorded 2026-09-22
loss_jocor algorithm/jocor.py ran · our draft was wrong no licence file found · pointer only · 2d8dfe0f2d9f22e2 · report fixture: signature y_1:Tensor/2/float, y_2:Tensor/2/float, t:Tensor/1/int, forget_rate:float/0/float, ind:ndarray/1/int, noise_or_not:ndarray/1/int, co_lambda:float/0/float · returned tuple · level recorded 2026-09-22
JoCoR algorithm/jocor.py unverified no licence file found · pointer only · 0fa67743b174d70a · report
hongxin001/open-sampling community (archive-listed) 9 samples, 2 ran (1 instrument-failed) · 7 unverified
NormedLinear alg/ood_noise.py ran · metamorphic tier: invariant no licence file found · pointer only · 4846ae16e0d273a9 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
resnet32 in 47 papers alg/ood_noise.py ran · our draft was wrong no licence file found · pointer only · 6c0f9be9596f3ff6 · report fixture: signature num_classes:int/0/int, use_norm:bool/0/bool · returned ResNet_s · level recorded 2026-09-22
BaseAlg alg/ood_noise.py unverified no licence file found · pointer only · 9860d0284775de1f · report
BasicBlock in 126 papers
census: 6 implementations from 6 papers, 5 distinct outputs on one shared input
alg/ood_noise.py unverified no licence file found · pointer only · afb150f3f055ddef · report
OEScratch alg/ood_noise.py unverified no licence file found · pointer only · ff341f03fb04bd64 · report
OODNoise alg/ood_noise.py unverified no licence file found · pointer only · 168776367e5649b5 · report
ResNet_s alg/ood_noise.py unverified no licence file found · pointer only · 2fdb5443e73b1ec6 · report
create_model in 163 papers alg/ood_noise.py unverified no licence file found · pointer only · 3a41d4c5de5de3f2 · report
create_ood_dataset alg/ood_noise.py unverified no licence file found · pointer only · 3f1cd084d493d247 · report
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.
call_bn in 5 papers repository not located ran · our draft was wrong licence of this copy not recorded · 7937f88152d63f61 · report fixture: signature bn:callable/0/?, x:Tensor/2/float · returned Tensor · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

3 assemblies on this paper: 3 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

3 sibling assemblies for Open-set samples with Dynamic Noisy Labels (ODNL) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Open-set samples with Dynamic Noisy Labels (ODNL) (graph method: Open-set samples with Dynamic Noisy Labels (ODNL)). 14 bricks; individually on a fixture: 10 ran (of which 6 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 1 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. ODNL entry · level 0 · unverified · b998044858807188
  2. SingleModel helper · level 1 · ran · 5ef095faf1a178df
  3. build_dataset data_transform · level 0 · unverified · b3a727f63915052f
  4. build_ood_noise helper · level 1 · ran · our draft was wrong · 6463c81eb9a79759
  5. build_model helper · level 1 · ran; the contract check could not run (DEP_MISSING) · 00092d7deaae3c5a
  6. CIFAR100 helper · level 0 · unverified · b83e460dc32b7006
  7. SVHN helper · level 1 · ran · a48cbe21ee4b47c7
  8. get_dataset_normlize helper · level 1 · ran · honoured contract · 331a7108385c91ba
  9. WideResNet helper · level 1 · ran · bbf25f185bcee0a2
  10. DependentLabelGenerator helper · level 1 · ran · 3aa8c19f6a93cc4f
  11. check_integrity helper · level 1 · ran · violated contract · 379a713ff6573163
  12. download_url helper · level 0 · unverified · 0105f183fe8b9666
  13. BasicBlock helper · level 1 · ran · 68c820cf9fac1d45
  14. NetworkBlock helper · level 1 · ran · dc96cf9441e16644

harvested Assembly for Open-set samples with Dynamic Noisy Labels (ODNL) (graph method: Open-set samples with Dynamic Noisy Labels (ODNL)). 7 bricks; individually on a fixture: 6 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; 4 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. JoCoR entry · level 0 · unverified · 0fa67743b174d70a
  2. CNN model · level 1 · ran · c15655fa5cf01c52
  3. MLPNet model · level 1 · ran · dca203cfebebf4d7
  4. accuracy helper · level 1 · ran · our draft was wrong · 5654f848f08debcf
  5. loss_jocor loss · level 1 · ran · our draft was wrong · 2d8dfe0f2d9f22e2
  6. call_bn helper · level 1 · ran · our draft was wrong · 7937f88152d63f61
  7. kl_loss_compute loss · level 1 · ran · our draft was wrong · 75feb4c4aa30068a

harvested Assembly for Open-set samples with Dynamic Noisy Labels (ODNL) (graph method: Open-set samples with Dynamic Noisy Labels (ODNL)). 16 bricks; individually on a fixture: 7 ran (of which 2 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) · 9 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. OODNoise entry · level 0 · unverified · 168776367e5649b5
  2. OEScratch helper · level 0 · unverified · ff341f03fb04bd64
  3. accuracy helper · level 1 · ran; the contract check could not run · 1e369e39c90cada6
  4. AverageMeter helper · level 1 · ran · 8c2fbaf0a3677a09
  5. BaseAlg helper · level 0 · unverified · 9860d0284775de1f
  6. create_ood_dataset helper · level 0 · unverified · 3f1cd084d493d247
  7. create_ood_noise helper · level 1 · ran · our draft was wrong · aa5465518a0ebdbf
  8. create_model helper · level 0 · unverified · 3a41d4c5de5de3f2
  9. TinyImages helper · level 0 · unverified · 6c523c58024ba795
  10. get_data_transform helper · level 1 · ran · our draft was wrong · 6526f339ad3f174a
  11. resnet32 helper · level 1 · ran · our draft was wrong · 6c0f9be9596f3ff6
  12. BasicBlock helper · level 0 · unverified · afb150f3f055ddef
  13. ResNet_s helper · level 0 · unverified · 2fdb5443e73b1ec6
  14. LambdaLayer helper · level 1 · ran · 17e20d525a2f3eb0
  15. NormedLinear helper · level 1 · ran · 4846ae16e0d273a9
  16. _weights_init helper · level 0 · unverified · 606494cf7b074c7e

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2106.10891") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Learning with noisy labelsOut-of-Distribution DetectionWeakly-supervised Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SGD

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