Papers › Unsupervised Data Augmentation for Consistency Training
Unsupervised Data Augmentation for Consistency Training
Qizhe Xie, Zihang Dai, Eduard Hovy, Minh-Thang Luong, Quoc V. Le
Semi-supervised learning lately has shown much promise in improving deep learning models when labeled data is scarce. Common among recent approaches is the use of consistency training on a large amount of unlabeled data to constrain model predictions to be invariant to input noise. In this work, we present a new perspective on how to effectively noise unlabeled examples and argue that the quality of noising, specifically those produced by advanced data augmentation methods, plays a crucial role in semi-supervised learning. By substituting simple noising operations with advanced data augmentation methods such as RandAugment and back-translation, our method brings substantial improvements across six language and three vision tasks under the same consistency training framework. On the IMDb text classification dataset, with only 20 labeled examples, our method achieves an error rate of 4.20, outperforming the state-of-the-art model trained on 25,000 labeled examples. On a standard semi-supervised learning benchmark, CIFAR-10, our method outperforms all previous approaches and achieves an error rate of 5.43 with only 250 examples. Our method also combines well with transfer learning, e.g., when finetuning from BERT, and yields improvements in high-data regime, such as ImageNet, whether when there is only 10% labeled data or when a full labeled set with 1.3M extra unlabeled examples is used. Code is available at https://github.com/google-research/uda.
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):
get_harvested_code_for_paper(arxiv_id="1904.12848")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="1904.12848", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="1904.12848")
Code
Syntology 52 code samples harvested from 12 repositories linked to this paper: 30 ran (of which 3 constructed an object rather than computing a result; 22 with no instrument failure: 2 honoured, 0 violated, 20 with no contract checked; 8 where Syntology's instrument failed) · 22 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 4 where the contract check could not run; failures of Syntology's instrument, not of the code.
By repository: official repository: 24 samples from 1 repository, 10 ran (10 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 28 samples from 11 repositories, 20 ran (12 with no instrument failure; 8 where Syntology's instrument failed). 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:
- uizard-technologies/realmix, community (archive-listed): harvested for another paper, listed under the repository
Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 30 samples that ran has no recorded date.
20 repositories listed; official and paper-mentioned ones first.
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
52 samples harvested: 30 ran (of which 3 constructed an object rather than computing a result; 22 with no instrument failure: 2 honoured, 0 violated, 20 with no contract checked; 8 where Syntology's instrument failed) · 22 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.
Licence: 17 of the 52 samples are pointer only, meaning Syntology does not serve that copy's text. For 8 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.
Harvested from 12 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. “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.
98430a824c0f53a8 · report
fixture: signature examples:list/1/Example · returned dict · level recorded 2026-09-25 a7723bf702941670 · report
fixture: signature img_height:int/0/int, img_width:int/0/int, num_channels:int/0/int, size:int/0/int · returned tuple · level recorded 2026-09-25 d00f20643f490779 · report
fixture: signature split:str/0/str, sup_size:int/0/int · returned str · level recorded 2026-09-25 5b70acc56bd3291f · report
fixture: signature split:str/0/str, sup_size:int/0/int · returned str · level recorded 2026-09-25 d90af0b30176ac15 · report
fixture: signature aug_copy_num:int/0/int · returned str · level recorded 2026-09-25 1902057ba4826690 · report
fixture: signature log_p:EagerTensor/2/float, log_q:EagerTensor/2/float · returned SymbolicTensor · level recorded 2026-09-25 6e1198a9c8c47438 · report
fixture: signature x:ndarray/2/float · returned ndarray · level recorded 2026-09-25 c6512991cdef483b · report
fixture: signature sent:str/0/str, punc:str/0/str, offset:int/0/int · returned list · level recorded 2026-09-25 563e25e514f4a2a6 · report
fixture: signature inputs:EagerTensor/4/float, in_filter:int/0/int, out_filter:int/0/int · returned EagerTensor · level recorded 2026-09-25 f3f5fce61e36ad06 · report
fixture: signature img:ndarray/3/float, amount:int/0/int · returned ndarray · level recorded 2026-09-25 914d0b8791ca87d0 · report
97c1d8a99764fb25 · report
d468fe223b1a25a2 · report
b417179a5b5e0e14 · report
89985ecf00256155 · report
17411729594afb42 · report
f6b18b79eccd8391 · report
36741a983d72bd68 · report
fixture: returned MagnitudePolicy · level recorded 2026-09-22 5722c1e755019115 · report
fixture: returned RandAugment · level recorded 2026-09-22 36c684d7fa0846c8 · report
bfec53e668f92651 · report
fixture: signature unsup_logits:Tensor/2/float, aug_logits:Tensor/2/float, criterion:callable/0/?, beta:float/0/float, teta:float/0/float · returned Tensor · level recorded 2026-09-22 1a1dcc06f81c3c75 · report
fixture: signature x:Image/0/? · returned Image · level recorded 2026-09-22 b9d05e119e47da9f · report
41b7d839c5008446 · report
3458a17eb1c6c95f · report
df213afe1b71adfd · report
fixture: signature examples:list/1/str, max_seq_length:int/0/int, tokenizer:SimpleTokenizer/0/?, output_mode:str/0/str, label_list:NoneType/0/? · returned list · level recorded 2026-09-22 b92936d1f2ac3e6c · report
fixture: signature img:Image/0/?, v:float/0/float · returned Image · level recorded 2026-09-22 008c64b77c4ae4c9 · report
fixture: signature img:Image/0/?, v:float/0/float · returned Image · level recorded 2026-09-22 ab50d5bbcf589623 · report
fixture: signature img:Image/0/?, v:float/0/float · returned Image · level recorded 2026-09-22 cbc36864ed1d2c42 · report
fixture: signature seq:list/1/int · returned list · level recorded 2026-09-24 00e569acd6b45ef0 · report
fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded, date not recorded f018413f261238f7 · report
fixture: signature input:Tensor/2/float, target:Tensor/1/int, size_average:bool/0/bool, label_smoothing:float/0/float · returned Tensor · level recorded 2026-09-24 406265b8de6d26a2 · report
fixture: signature arr:list/1/int · returned list · level recorded 2026-09-24 40d3c39a403ae1f7 · report
fixture: signature level:int/0/int, maxval:float/0/float · returned float · level recorded 2026-09-24 6066adc7a58afde6 · report
fixture: signature name:str/0/str, level:int/0/int · returned Logger · level recorded 2026-09-25 7900a3a6deef908b · report
fixture: signature level:int/0/int, maxval:int/0/int · returned int · level recorded 2026-09-24 6cb91780680fb891 · report
fixture: signature songs_list:list/1/str, artists_list:list/1/str · returned list · level recorded 2026-09-24 census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket metrics.py unverified Apache-2.0 (permissive) ·
3ccb35170e45f022 · report
64a6c97a5657e37a · report
fixture: signature ori_text:str/0/str, new_text:str/0/str, use_min_length:int/0/int, use_max_length_diff_ratio:float/0/float · returned str · level recorded 2026-09-22 2f093dc1c7296f6c · report
9355efdc819d943e · report
fixture: returned AverageMeter · level recorded 2026-09-22 a359f54f3b9012c2 · report
fixture: returned float · level recorded 2026-09-22 e69c03b0ebf8765a · report
fixture: signature model:callable/0/?, x:Tensor/2/float, x_h:Tensor/2/float · returned Tensor · level recorded 2026-09-24 83a7585173c63bc5 · report
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.
11 assemblies on this paper: 11 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).
11 sibling assemblies for Unsupervised Data Augmentation for Consistency Training (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 1 brick; individually on a fixture: 1 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; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- get_unsupervised_loss entry · level 1 · ran · our draft was wrong ·
bfec53e668f92651
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 2 bricks; individually on a fixture: 2 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; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- uda_train entry · level 1 · ran; the contract check could not run (RAISES) ·
a359f54f3b9012c2 - AverageMeter helper · level 1 · ran ·
9355efdc819d943e
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- jensen_shannon_divergence_consistency_loss entry · level 0 · unverified ·
36c684d7fa0846c8
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 1 brick; individually on a fixture: 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. The assembly has not been run end-to-end.
one of 11 siblings
- UDALoss entry · level 1 · ran ·
e69c03b0ebf8765a
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 2 bricks; individually on a fixture: 1 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; 1 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- back_translation entry · level 0 · unverified ·
2f093dc1c7296f6c - replace_with_length_check data_transform · level 1 · ran · our draft was wrong ·
64a6c97a5657e37a
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 20 bricks; individually on a fixture: 20 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; 20 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- augment_list entry · level 1 · ran · our draft was wrong ·
b1cd4e4b4bba7f8c - AutoContrast data_transform · level 1 · ran · our draft was wrong ·
4a0ac9d611e018bc - Brightness data_transform · level 1 · ran; the contract check could not run ·
993134d3c95bcf92 - Color data_transform · level 1 · ran; the contract check could not run ·
ccaa14b841bf44dc - Contrast data_transform · level 1 · ran; the contract check could not run ·
35f5a6d3913e6b0e - Cutout data_transform · level 1 · ran; the contract check could not run ·
f90119cfdd38ef56 - CutoutAbs helper · level 1 · ran; the contract check could not run ·
03ecbd868107e6f7 - Equalize data_transform · level 1 · ran · our draft was wrong ·
4befb1744e90f3a1 - Invert data_transform · level 1 · ran · our draft was wrong ·
f5f0bf976dfcf068 - Posterize data_transform · level 1 · ran; the contract check could not run ·
d2a9bd74240b6944 - Posterize2 helper · level 1 · ran; the contract check could not run ·
067680c1cc9f4b97 - Rotate data_transform · level 1 · ran · our draft was wrong ·
21c6427b09f14649 - Sharpness data_transform · level 1 · ran; the contract check could not run ·
e1ccc58bf76a30b9 - ShearX data_transform · level 1 · ran; the contract check could not run (RAISES) ·
b92936d1f2ac3e6c - ShearY data_transform · level 1 · ran; the contract check could not run (RAISES) ·
008c64b77c4ae4c9 - Solarize data_transform · level 1 · ran; the contract check could not run ·
70eab251c4b2bce2 - TranslateX data_transform · level 1 · ran; the contract check could not run (RAISES) ·
ab50d5bbcf589623 - TranslateXAbs helper · level 1 · ran; the contract check could not run ·
b1fc55be13be06e7 - TranslateY data_transform · level 1 · ran; the contract check could not run ·
d4b73cc9fa9b08ae - TranslateYAbs helper · level 1 · ran; the contract check could not run ·
15ceb820d17f0a5c
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 2 bricks; individually on a fixture: 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. The assembly has not been run end-to-end.
one of 11 siblings
- RandAugment entry · level 1 · ran ·
5722c1e755019115 - MagnitudePolicy helper · level 1 · ran ·
36741a983d72bd68
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- get_uda_classification_loss entry · level 0 · unverified ·
83a7585173c63bc5
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 2 bricks; individually on a fixture: 2 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; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- convert_examplesUDA_to_features entry · level 1 · ran · our draft was wrong ·
df213afe1b71adfd - InputFeatures helper · level 1 · ran ·
a8d8657df9ba7457
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 10 bricks; individually on a fixture: 9 ran (of which 4 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 0 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- Classifier entry · level 0 · unverified ·
e46d15e93318b254 - Transformer model · level 1 · ran ·
4076b007bcd248ae - Block helper · level 1 · ran ·
2dcf1b4f0aa2fa9a - Embeddings model · level 1 · ran ·
b036f3e72b73df3d - LayerNorm model · level 1 · ran ·
b0f94b02a9a6c057 - MultiHeadedSelfAttention model · level 1 · ran ·
05a7a0d5018b97f8 - PositionWiseFeedForward model · level 1 · ran ·
a10a2edb3945c395 - merge_last helper · level 1 · ran; the contract check could not run ·
d5e69fc7b0ee7c8c - split_last helper · level 1 · ran; the contract check could not run ·
a4a5354a3138c8ef - gelu helper · level 1 · ran · honoured contract ·
62e7409a816626ac
harvested Assembly for Unsupervised Data Augmentation for Consistency Training (graph method: Unsupervised Data Augmentation for Consistency Training). 4 bricks; individually on a fixture: 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) · 3 unverified. The assembly has not been run end-to-end.
one of 11 siblings
- StrongAugmentation entry · level 0 · unverified ·
3458a17eb1c6c95f - RGB2RGBA helper · level 1 · ran ·
1a1dcc06f81c3c75 - RandAugment helper · level 0 · unverified ·
b9d05e119e47da9f - ReduceChannelwithNormalize helper · level 0 · unverified ·
41b7d839c5008446
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="1904.12848") returns these assemblies with each brick's role and wiring (how to connect).
Tasks
Results from the paper archive 2025-07-28
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
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