Papers › RoBERTa: A Robustly Optimized BERT Pretraining Approach

RoBERTa: A Robustly Optimized BERT Pretraining Approach

26 Jul 2019arXiv:1907.11692archive 2025-07-28

Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov

Language model pretraining has led to significant performance gains but careful comparison between different approaches is challenging. Training is computationally expensive, often done on private datasets of different sizes, and, as we will show, hyperparameter choices have significant impact on the final results. We present a replication study of BERT pretraining (Devlin et al., 2019) that carefully measures the impact of many key hyperparameters and training data size. We find that BERT was significantly undertrained, and can match or exceed the performance of every model published after it. Our best model achieves state-of-the-art results on GLUE, RACE and SQuAD. These results highlight the importance of previously overlooked design choices, and raise questions about the source of recently reported improvements. We release our models and code.

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

No sample from the archive's official repository (pytorch/fairseq) is linked to this paper; the 48 samples below come from 16 community repositories. Syntology harvested code from pytorch/fairseq for 3 other papers.

pytorch/fairseq: Syntology's graph links these samples to LEWIS: Levenshtein Editing for Unsupervised Text Style Transfer (arXiv:2105.08206; 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), Cascaded Head-colliding Attention (arXiv:2105.14850; that paper's own run record: 9 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 1 violated, 2 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified) and Modeling Dual Read/Write Paths for Simultaneous Machine Translation (arXiv:2203.09163; that paper's own run record: 4 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; 2 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method.

Syntology 48 code samples harvested from 16 repositories linked to this paper: 37 ran (of which 11 constructed an object rather than computing a result; 36 with no instrument failure: 2 honoured, 0 violated, 34 with no contract checked; 1 where Syntology's instrument failed) · 11 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 48 samples from 16 repositories, 37 ran (36 with no instrument failure; 1 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 14 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

67 repositories listed; official and paper-mentioned ones first.

pytorch/fairseq officialmentioned in paperpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LEWIS: Levenshtein Editing for Unsupervised Text Style Transfer (arXiv:2105.08206; 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), Cascaded Head-colliding Attention (arXiv:2105.14850; that paper's own run record: 9 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 1 violated, 2 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified) and Modeling Dual Read/Write Paths for Simultaneous Machine Translation (arXiv:2203.09163; that paper's own run record: 4 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; 2 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
10Exahertz/Text-ResNet-on-Sentimental-LIAR-Fake-News mentioned on GitHubtfSyntology: not harvested report
CalumPerrio/WNUT-2020 mentioned on GitHubpytorchGPL-3.0Syntology: not harvested report
G-4-R-Y/Tweet-Sentiment-Extraction mentioned on GitHubSyntology: not harvested report
G-4-R-Y/Tweet-Sentiment-Extraction-roBERTa-5fold mentioned on GitHubSyntology: not harvested report
GeorgeLuImmortal/Hierarchical-BERT-Model-with-Limited-Labelled-Data mentioned on GitHubpytorchSyntology: 10 samples, 8 ran (0 instrument-failed) · 2 unverified report
IndicoDataSolutions/finetune mentioned on GitHubtfMPL-2.0Syntology: not harvested report
Karthik-Bhaskar/Context-Based-Question-Answering mentioned on GitHubtfApache-2.0Syntology: not harvested report
NathanDuran/Sentence-Encoding-for-DA-Classification mentioned on GitHubtfGPL-3.0Syntology: not harvested report
SindhuMadi/FakeNewsDetection mentioned on GitHubSyntology: not harvested report
Tencent/TurboTransformers mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
UnknownGenie/altered-BERT-KPE mentioned on GitHubpytorchSyntology: not harvested report
abdumaa/hiqualprop mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
abhishekanand1710/noiseandbias mentioned on GitHubSyntology: not harvested report
aistairc/kirt_bert_on_abci mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
awslabs/mlm-scoring mentioned on GitHubmxnetSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (arXiv:1810.04805; that paper's own run record: 300 ran (of which 75 constructed an object rather than computing a result; 234 with no instrument failure: 18 honoured, 5 violated, 211 with no contract checked; 66 where Syntology's instrument failed) · 359 unverified); nothing checks that they implement this paper's method. report
bcaitech1/p2-klue-Heeseok-Jeong mentioned on GitHubpytorchSyntology: not harvested report
benywon/ReCO mentioned on GitHubpytorchSyntology: not harvested report
bfopengradient/NLP_ROBERTA mentioned on GitHubSyntology: not harvested report
blawok/named-entity-recognition mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
bluejurand/Kaggle_QA_Google_Labeling mentioned on GitHubtfSyntology: not harvested report
brightmart/roberta_zh mentioned on GitHubtfSyntology: not harvested report
clovaai/textual-kd-slu mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations (arXiv:1910.05453; that paper's own run record: 3 ran (of which 0 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) · 0 unverified); nothing checks that they implement this paper's method. report
common-english/bert-all mentioned on GitHubpytorchSyntology: not harvested report
devhemza/BERTweet_sentiment_analysis mentioned on GitHubpytorchSyntology: not harvested report
dig-team/hanna-benchmark-asg mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to CTRL: A Conditional Transformer Language Model for Controllable Generation (arXiv:1909.05858; that paper's own run record: 14 ran (of which 3 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 1 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
duanchi1230/NLP_Project_AI2_Reasoning_Challenge mentioned on GitHubpytorchSyntology: not harvested report
eternityyw/tram-benchmark mentioned on GitHubMITSyntology: not harvested report
expertailab/spaceqa mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
facebookresearch/anli mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
few-shot-NER-benchmark/BaselineCode mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
flexible-fl/flex-nlp mentioned on GitHubAGPL-3.0Syntology: not harvested report
haisongzhang/roberta-tiny-cased mentioned on GitHubSyntology: not harvested report
hkuds/easyrec mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
huggingface/transformers mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified), HuggingFace's Transformers: State-of-the-art Natural Language Processing (arXiv:1910.03771; that paper's own run record: 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) · 0 unverified), DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
ibm/vira-intent-discovery mentioned on GitHubApache-2.0Syntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
kaushaltrivedi/fast-bert mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
knuddj1/op_text mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter (arXiv:1910.01108; that paper's own run record: 21 ran (of which 5 constructed an object rather than computing a result; 12 with no instrument failure: 3 honoured, 1 violated, 8 with no contract checked; 9 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
knuddy/op_text mentioned on GitHubpytorchMITSyntology: not harvested report
lashoun/hanna-benchmark-asg mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Of Human Criteria and Automatic Metrics: A Benchmark of the Evaluation of Story Generation (arXiv:2208.11646; that paper's own run record: 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); nothing checks that they implement this paper's method. report
mthcom/hscore-dataset-pruning mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to OPT: Open Pre-trained Transformer Language Models (arXiv:2205.01068; that paper's own run record: 14 ran (of which 2 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 1 where Syntology's instrument failed) · 10 unverified); nothing checks that they implement this paper's method. report
musixmatchresearch/umberto mentioned on GitHubpytorchMITSyntology: not harvested report
nguyenvulebinh/vietnamese-roberta mentioned on GitHubpytorchSyntology: not harvested report
obi-ml-public/ehr_deidentification mentioned on GitHubMITSyntology: not harvested report
octanove/shiba mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
oneflow-inc/libai mentioned on GitHubApache-2.0Syntology: not harvested report
pisalore/roberta_results mentioned on GitHubpytorchMITSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
ricaelum42/Contextual-Twitter-Sarcasm-Detection mentioned on GitHubpytorchSyntology: not harvested report
salesforce/codet5 mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
sdadas/polish-roberta mentioned on GitHubpytorchLGPL-3.0Syntology: not harvested report
simon-benigeri/narrative-generation mentioned on GitHubpytorchSyntology: not harvested report
tighu20/Kaggle-Tweet-Sentiment-Extraction mentioned on GitHubtfSyntology: not harvested report
traviscoan/cards mentioned on GitHubApache-2.0Syntology: 8 samples, 7 ran (0 instrument-failed) · 1 unverified report
utterworks/fast-bert mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to XLNet: Generalized Autoregressive Pretraining for Language Understanding (arXiv:1906.08237; that paper's own run record: 15 ran (of which 2 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 9 unverified); nothing checks that they implement this paper's method. report
viethoang1512/kpa mentioned on GitHubpytorchApache-2.0Syntology: 4 samples, 4 ran (0 instrument-failed) · 0 unverified report
wzzzd/LM_NER mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
xiaoqian19940510/text-classification- mentioned on GitHubpytorchSyntology: not harvested report
xiaoqian19940510/text-classification-surveys mentioned on GitHubpytorchSyntology: 6 samples, 6 ran (0 instrument-failed) · 0 unverified report
zaradana/Fast_BERT mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
zfj1998/CodeBert-Code2Text mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to CodeBERT: A Pre-Trained Model for Programming and Natural Languages (arXiv:2002.08155; that paper's own run record: 14 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
znhy1024/protoco mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to OPT: Open Pre-trained Transformer Language Models (arXiv:2205.01068; that paper's own run record: 14 ran (of which 2 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 1 where Syntology's instrument failed) · 10 unverified); nothing checks that they implement this paper's method. report
2023-MindSpore-1/ms-code-163 mindsporeSyntology: not harvested report
MS-P3/code7 mindsporeSyntology: not harvested report
PaddlePaddle/PaddleNLP paddleSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ERNIE 3.0 Tiny: Frustratingly Simple Method to Improve Task-Agnostic Distillation... (arXiv:2301.03416; that paper's own run record: 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) · 1 unverified); nothing checks that they implement this paper's method. report
lvyufeng/bert4ms mindsporeApache-2.0Syntology: 4 samples, 2 ran (0 instrument-failed) · 2 unverified report
pwc-1/Paper-9 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RandAugment: Practical automated data augmentation with a reduced search space (arXiv:1909.13719; that paper's own run record: 58 ran (of which 1 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 51 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/University mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution (arXiv:2409.12191; that paper's own run record: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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

48 samples harvested: 37 ran (of which 11 constructed an object rather than computing a result; 36 with no instrument failure: 2 honoured, 0 violated, 34 with no contract checked; 1 where Syntology's instrument failed) · 11 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.

2ran · honoured contract
1ran · our draft was wrong
34ran
11unverified

Licence: 24 of the 48 samples are pointer only, meaning Syntology does not serve that copy's text. For 1 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 16 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”. For 7 samples 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.

GeorgeLuImmortal/Hierarchical-BERT-Model-with-Limited-Labelled-Data community (archive-listed) 10 samples, 8 ran (0 instrument-failed) · 2 unverified
BertAttention in 28 papers run_hbm.py ran · metamorphic tier: deterministic no licence file found · pointer only · ad7f6e57dedf08aa · report fixture: returned BertAttention · level recorded 2026-09-22
BertEmbeddings in 21 papers run_hbm.py ran fingerprinted no licence file found · pointer only · c3af34b4f438eaea · report fixture: signature inputs_embeds:Tensor/3/float, position_ids:NoneType/0/? · returned Tensor · level recorded 2026-09-24
BertIntermediate in 15 papers run_hbm.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 6e60ffe93c5ec5fa · report fixture: returned BertIntermediate · level recorded 2026-09-22
BertModel in 16 papers run_hbm.py ran fingerprinted no licence file found · pointer only · f43b4161b7bdba24 · report fixture: signature inputs_embeds:Tensor/3/float, attention_mask:Tensor/2/float · returned tuple · level recorded 2026-09-24
BertOutput in 13 papers run_hbm.py ran no licence file found · pointer only · 5e5629cb8e93e004 · report fixture: signature hidden_states:Tensor/2/float, input_tensor:Tensor/2/float · returned Tensor · level recorded 2026-09-24
BertPooler in 6 papers run_hbm.py ran · metamorphic tier: deterministic no licence file found · pointer only · 4e755eff48bf0752 · report fixture: returned BertPooler · level recorded 2026-09-22
BertSelfAttention in 29 papers run_hbm.py ran fingerprinted no licence file found · pointer only · df8363b5430892a0 · report fixture: signature hidden_states:Tensor/3/float, attention_mask:NoneType/0/? · returned tuple · level recorded 2026-09-24
BertSelfOutput in 14 papers run_hbm.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · fcf3cd282eb03efe · report fixture: returned BertSelfOutput · level recorded 2026-09-22
BertEncoder in 20 papers run_hbm.py unverified no licence file found · pointer only · 00ea7705507db813 · report
BertLayer in 23 papers run_hbm.py unverified no licence file found · pointer only · f2ab856479b90e8a · report
abdumaa/hiqualprop community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
ClassificationHead in 12 papers full/custom_sequence_classification.py ran · metamorphic tier: deterministic no licence file found · pointer only · 7b8d60647337c2b1 · report fixture: returned ClassificationHead · level recorded 2026-09-22
CustomRobertaForSequenceClassification full/custom_sequence_classification.py ran no licence file found · pointer only · 0196b04010ddfa00 · report fixture: signature input_ids:Tensor/2/int, attention_mask:Tensor/2/float, extra_data:Tensor/2/float, labels:Tensor/1/int · returned SequenceClassifierOutput · level recorded 2026-09-24
blawok/named-entity-recognition community (archive-listed) 1 sample, 0 ran · 1 unverified
train in 522 papers src/source_roberta/train_roberta.py unverified MIT (permissive) · 92e80bcaa1f41216 · report
expertailab/spaceqa community (archive-listed) 1 sample, 0 ran · 1 unverified
QuestionAnswering question_answering/src/question_answering/question_answering.py unverified Apache-2.0 (permissive) · da835c893505251b · report
few-shot-NER-benchmark/BaselineCode community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
RobertaNER src/meta_model.py ran no licence file found · pointer only · 152609ff366dfb9e · report fixture: signature input_ids:Tensor/2/int, support_class_num:int/0/int, orig_prototypes:NoneType/0/?, orig_embed_class_len:NoneType/0/?, attention_mask:Tensor/2/float, position_ids:NoneType/0/?, head_mask:NoneType/0/?, inputs_embeds:NoneType/0/?, labels:Tensor/2/int · returned tuple · level recorded 2026-09-24
hkuds/easyrec community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
Pooler in 8 papers model.py ran · metamorphic tier: deterministic no licence file found · pointer only · dba0b61c9d62f82f · report fixture: returned Pooler · level recorded 2026-09-22
Easyrec model.py unverified no licence file found · pointer only · b2bd29f8a81790e2 · report
ibm/vira-intent-discovery community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
convert_to_single_none_row_df trim.py ran Apache-2.0 (permissive) · e06455e72782e6d0 · report fixture: signature none_df:DataFrame/0/? · returned DataFrame · level recorded 2026-09-24
filter_all_data prepare.py ran Apache-2.0 (permissive) · d4db5e37c5831f1a · report fixture: signature row:dict/0/? · returned bool · level recorded 2026-09-24
filter_oracle_train prepare.py ran Apache-2.0 (permissive) · 57b444571a95cb8a · report fixture: signature row:dict/0/? · returned bool · level recorded 2026-09-24
kaushaltrivedi/fast-bert community (archive-listed) 1 sample, 0 ran · 1 unverified
RobertaForMultiLabelSequenceClassification fast_bert/modeling.py unverified Apache-2.0 (permissive) · b3e0c37e1e2b9503 · report
pisalore/roberta_results community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
get_tasks in 21 papers glue_tasks/download_glue_data.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 3caf26b311c98658 · report fixture: signature task_names:str/0/str · returned list · level recorded 2026-09-22
salesforce/codet5 community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
RobertaClassificationHead CodeT5/models.py ran · metamorphic tier: invariant fingerprinted BSD-3-Clause (permissive) · 17243795c9ab1ec3 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
traviscoan/cards community (archive-listed) 8 samples, 7 ran (0 instrument-failed) · 1 unverified
drop_duplicates utils.py ran fingerprinted Apache-2.0 (permissive) · 163e38293198fa6a · report fixture: signature seq:list/1/int · returned list · level recorded 2026-09-24
estimate_ensemble fit/ensemble.py ran Apache-2.0 (permissive) · 9f11fd760e57188c · report fixture: signature probs1:ndarray/2/float, probs2:ndarray/2/float, le:LabelEncoder/0/? · returned list · level recorded 2026-09-24
flatten in 178 papers
census: 4 implementations from 7 papers, 4 distinct outputs across 4 buckets, one shared input per bucket
utils.py ran Apache-2.0 (permissive) · 954d7736e2d968f3 · report fixture: signature list_of_lists:list/1/list · returned list · level recorded 2026-09-24
predict_claim fit/ensemble.py ran Apache-2.0 (permissive) · 712ccc81e62f75b9 · report fixture: signature probs:ndarray/2/float, labels:list/1/str · returned list · level recorded 2026-09-24
read_csv in 60 papers utils.py ran Apache-2.0 (permissive) · df853d8826e7a476 · report fixture: signature path:str/0/str, remove_header:bool/0/bool · returned list · level recorded 2026-09-24
remove_between_square_brackets preprocess.py ran fingerprinted Apache-2.0 (permissive) · cd2c2c1263e0882d · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-24
remove_non_ascii preprocess.py ran fingerprinted Apache-2.0 (permissive) · a9831bf0d050a65c · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-24
fit_logistic_classifier fit/logistic.py unverified Apache-2.0 (permissive) · 53e20baefb5f4cd1 · report
viethoang1512/kpa community (archive-listed) 4 samples, 4 ran (0 instrument-failed) · 0 unverified
contrastive_loss in 80 papers
census: 11 implementations from 18 papers, 7 distinct outputs across 3 buckets, one shared input per bucket
qs_kpa/baselines/tf_models.py ran Apache-2.0 (permissive) · a5f2db749ea77f58 · report fixture: signature y:EagerTensor/1/float, preds:EagerTensor/1/float, margin:float/0/float · returned EagerTensor · level recorded 2026-09-24
convert_label_to_similarity qs_kpa/losses/circle_loss.py ran fingerprinted Apache-2.0 (permissive) · 6708df0bbbc13e34 · report fixture: signature normed_feature:Tensor/2/float, label:Tensor/1/int · returned tuple · level recorded 2026-09-24
cos_sim in 34 papers
census: 5 implementations from 5 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
qs_kpa/losses/multiple_negatives_ranking_loss.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 8ca90d5388372be5 · report fixture: signature a:Tensor/2/float, b:Tensor/2/float · returned Tensor · level recorded 2026-09-24
dot_score in 10 papers qs_kpa/losses/multiple_negatives_ranking_loss.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · afa18317f44e63ad · report fixture: signature a:Tensor/2/float, b:Tensor/2/float · returned Tensor · level recorded 2026-09-24
wzzzd/LM_NER community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
CRF model/roberta_crf.py ran no licence file found · pointer only · e65cfdc18cd8c48f · report fixture: signature emissions:Tensor/3/float, tags:Tensor/2/int, mask:Tensor/2/int, reduction:str/0/str · returned Tensor · level recorded 2026-09-24
RoBertaCRF model/roberta_crf.py unverified no licence file found · pointer only · 7d40208cd5d0da23 · report
xiaoqian19940510/text-classification-surveys community (archive-listed) 6 samples, 6 ran (0 instrument-failed) · 0 unverified
Encoder in 278 papers ChineseTextClassification/models/Transformer.py ran no licence file found · pointer only · 68faf3c1193d95e7 · report fixture: returned Encoder · level recorded 2026-09-22
Model in 227 papers ChineseTextClassification/models/Transformer.py ran no licence file found · pointer only · 9d6930c9ef7f273b · report fixture: returned Model · level recorded 2026-09-22
Multi_Head_Attention ChineseTextClassification/models/Transformer.py ran no licence file found · pointer only · 717e1f9d4eae4141 · report fixture: returned Multi_Head_Attention · level recorded 2026-09-22
Position_wise_Feed_Forward ChineseTextClassification/models/Transformer.py ran fingerprinted no licence file found · pointer only · c15e979783157fb0 · report fixture: returned Position_wise_Feed_Forward · level recorded 2026-09-22
Positional_Encoding ChineseTextClassification/models/Transformer.py ran fingerprinted no licence file found · pointer only · ce8217cf9489c436 · report fixture: returned Positional_Encoding · level recorded 2026-09-22
Scaled_Dot_Product_Attention ChineseTextClassification/models/Transformer.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · efea4e3ee53c36c9 · report fixture: signature Q:Tensor/3/float, K:Tensor/3/float, V:Tensor/3/float, scale:float/0/float · returned Tensor · level recorded 2026-09-22
zaradana/Fast_BERT community (archive-listed) 1 sample, 0 ran · 1 unverified
RobertaForMultiLabelSequenceClassification 0_fast_bert/modeling.py unverified Apache-2.0 (permissive) · 6788010bd752077c · report
lvyufeng/bert4ms community (archive-listed) 4 samples, 2 ran (0 instrument-failed) · 2 unverified
convert_state_dict in 13 papers cybertron/utils.py ran Apache-2.0 (permissive) · 8afaa985063f723e · report fixture: signature module_or_pth_file:callable/0/?, model:str/0/str · returned dict · level recorded 2026-09-24
invert_mask in 14 papers cybertron/models/bart.py ran Apache-2.0 (permissive) · e8ae42ea576c0232 · report fixture: signature attention_mask:Tensor/2/int · returned Tensor · level recorded 2026-09-24
load_from_cache cybertron/utils.py unverified Apache-2.0 (permissive) · e462fa293bb314aa · 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.

12 assemblies on this paper: 12 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

12 sibling assemblies for RoBERTa (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for RoBERTa (graph method: RoBERTa). 4 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) · 2 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 12 siblings

  1. Easyrec entry · level 0 · unverified · b2bd29f8a81790e2
  2. MLPLayer model · level 0 · unverified · 7d7789da5378c84b
  3. Pooler model · level 1 · ran · dba0b61c9d62f82f
  4. Similarity helper · level 1 · ran · 4d41003f996241f8

harvested Assembly for RoBERTa (graph method: RoBERTa). 2 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) · 1 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. RoBertaCRF entry · level 0 · unverified · 7d40208cd5d0da23
  2. CRF model · level 1 · ran · e65cfdc18cd8c48f

harvested Assembly for RoBERTa (graph method: RoBERTa). 2 bricks; individually on a fixture: 2 ran (of which 1 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. The assembly has not been run end-to-end.

one of 12 siblings

  1. CustomRobertaForSequenceClassification entry · level 1 · ran · 0196b04010ddfa00
  2. ClassificationHead model · level 1 · ran · 7b8d60647337c2b1

harvested Assembly for RoBERTa (graph method: RoBERTa). 10 bricks; individually on a fixture: 8 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. BertModel entry · level 1 · ran · f43b4161b7bdba24
  2. BertEmbeddings model · level 1 · ran · c3af34b4f438eaea
  3. BertEncoder model · level 0 · unverified · 00ea7705507db813
  4. BertPooler model · level 1 · ran · 4e755eff48bf0752
  5. BertLayer model · level 0 · unverified · f2ab856479b90e8a
  6. BertAttention model · level 1 · ran · ad7f6e57dedf08aa
  7. BertIntermediate model · level 1 · ran · 6e60ffe93c5ec5fa
  8. BertOutput model · level 1 · ran · 5e5629cb8e93e004
  9. BertSelfAttention model · level 1 · ran · df8363b5430892a0
  10. BertSelfOutput model · level 1 · ran · fcf3cd282eb03efe

harvested Assembly for RoBERTa (graph method: RoBERTa). 6 bricks; individually on a fixture: 6 ran (of which 5 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) · 0 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. Model entry · level 1 · ran · 9d6930c9ef7f273b
  2. Encoder helper · level 1 · ran · 68faf3c1193d95e7
  3. Positional_Encoding model · level 1 · ran · ce8217cf9489c436
  4. Multi_Head_Attention model · level 1 · ran · 717e1f9d4eae4141
  5. Position_wise_Feed_Forward model · level 1 · ran · c15e979783157fb0
  6. Scaled_Dot_Product_Attention model · level 1 · ran · efea4e3ee53c36c9

harvested Assembly for RoBERTa (graph method: RoBERTa). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. QuestionAnswering entry · level 0 · unverified · da835c893505251b

harvested Assembly for RoBERTa (graph method: RoBERTa). 4 bricks; individually on a fixture: 4 ran (of which 3 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) · 0 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. Encoder entry · level 1 · ran · 68faf3c1193d95e7
  2. Multi_Head_Attention model · level 1 · ran · 717e1f9d4eae4141
  3. Position_wise_Feed_Forward model · level 1 · ran · c15e979783157fb0
  4. Scaled_Dot_Product_Attention helper · level 1 · ran · efea4e3ee53c36c9

harvested Assembly for RoBERTa (graph method: RoBERTa). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. RobertaForMultiLabelSequenceClassification entry · level 0 · unverified · b3e0c37e1e2b9503

harvested Assembly for RoBERTa (graph method: RoBERTa). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. RobertaForMultiLabelSequenceClassification entry · level 0 · unverified · 6788010bd752077c

harvested Assembly for RoBERTa (graph method: RoBERTa). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 12 siblings

  1. train entry · level 0 · unverified · 92e80bcaa1f41216

harvested Assembly for RoBERTa (graph method: RoBERTa). 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 12 siblings

  1. RobertaNER entry · level 1 · ran · 152609ff366dfb9e

harvested Assembly for RoBERTa (graph method: RoBERTa). 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 12 siblings

  1. RobertaClassificationHead entry · level 1 · ran · 17243795c9ab1ec3

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="1907.11692") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Common Sense ReasoningDocument Image ClassificationLanguage ModelingLanguage ModellingLexical SimplificationLinguistic AcceptabilityMulti-task Language UnderstandingNatural Language InferenceOnly Connect Walls Dataset Task 1 (Grouping)Question AnsweringReading ComprehensionRiddle SenseSemantic Textual SimilaritySentence CompletionSentiment AnalysisStock Market PredictionText ClassificationType prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Common Sense Reasoning CommonsenseQA RoBERTa-Large 355M Accuracy 72.1 #19 of 38 Archive leaderboard report
Common Sense Reasoning SWAG RoBERTa Test 89.9 #2 of 5 Archive leaderboard report
Document Image Classification RVL-CDIP Roberta base Accuracy 90.06 #31 of 31 Archive leaderboard report
Document Image Classification RVL-CDIP Roberta base Parameters 125M #31 of 31 Archive leaderboard report
Linguistic Acceptability CoLA RoBERTa (ensemble) Accuracy 67.8% #20 of 43 Archive leaderboard report
Natural Language Inference ANLI test RoBERTa (Large) A1 72.4 #5 of 25 Archive leaderboard report
Natural Language Inference ANLI test RoBERTa (Large) A2 49.8 #5 of 25 Archive leaderboard report
Natural Language Inference ANLI test RoBERTa (Large) A3 44.4 #5 of 25 Archive leaderboard report
Natural Language Inference MultiNLI RoBERTa Matched 90.8 #9 of 67 Archive leaderboard report
Natural Language Inference MultiNLI RoBERTa (ensemble) Mismatched 90.2 #60 of 67 Archive leaderboard report
Natural Language Inference QNLI RoBERTa (ensemble) Accuracy 98.9% #5 of 43 Archive leaderboard report
Natural Language Inference RTE RoBERTa Accuracy 88.2% #18 of 90 Archive leaderboard report
Natural Language Inference RTE RoBERTa (ensemble) Accuracy 88.2% #19 of 90 Archive leaderboard report
Natural Language Inference WNLI RoBERTa (ensemble) Accuracy 89 #9 of 23 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW RoBERTa (LARGE) # Correct Groups 29 ± 3 #22 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW RoBERTa (LARGE) # Solved Walls 0 ± 0 #22 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW RoBERTa (LARGE) Adjusted Mutual Information (AMI) 9.4 ± .4 #22 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW RoBERTa (LARGE) Adjusted Rand Index (ARI) 8.4 ± .3 #22 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW RoBERTa (LARGE) Fowlkes Mallows Score (FMS) 26.7 ± .2 #22 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW RoBERTa (LARGE) Wasserstein Distance (WD) 88.4 ± .4 #22 of 22 Archive leaderboard report
Question Answering PIQA RoBERTa-Large 355M Accuracy 79.4 #37 of 67 Archive leaderboard report
Question Answering Quora Question Pairs RoBERTa (ensemble) Accuracy 90.2% #6 of 19 Archive leaderboard report
Question Answering SIQA RoBERTa-Large 355M (fine-tuned) Accuracy 76.7 #12 of 24 Archive leaderboard report
Question Answering SQuAD2.0 RoBERTa (single model) EM 86.820 #90 of 286 Archive leaderboard report
Question Answering SQuAD2.0 RoBERTa (single model) F1 89.795 #90 of 286 Archive leaderboard report
Question Answering SQuAD2.0 dev RoBERTa (no data aug) EM 86.5 #3 of 13 Archive leaderboard report
Question Answering SQuAD2.0 dev RoBERTa (no data aug) F1 89.4 #3 of 13 Archive leaderboard report
Reading Comprehension RACE RoBERTa Accuracy 83.2 #7 of 24 Archive leaderboard report
Reading Comprehension RACE RoBERTa Accuracy (High) 81.3 #7 of 24 Archive leaderboard report
Reading Comprehension RACE RoBERTa Accuracy (Middle) 86.5 #7 of 24 Archive leaderboard report
Semantic Textual Similarity MRPC RoBERTa (ensemble) Accuracy 92.3% #3 of 45 Archive leaderboard report
Semantic Textual Similarity STS Benchmark RoBERTa Pearson Correlation 0.922 #7 of 66 Archive leaderboard report
Sentence Completion HellaSwag RoBERTa-Large Ensemble Accuracy 85.5 #24 of 89 Archive leaderboard report
Sentence Completion HellaSwag RoBERTa-Large 355M Accuracy 81.7 #40 of 89 Archive leaderboard report
Sentiment Analysis SST-2 Binary classification RoBERTa (ensemble) Accuracy 96.7 #13 of 87 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News+Factors) Accuray 62.49 #11 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News+Factors) F1-score 62.54 #11 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News+Factors) Precision 62.59 #11 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News+Factors) Recall 62.51 #11 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News) Accuray 61.34 #12 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News) F1-score 61.48 #12 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News) Precision 61.97 #12 of 17 Archive leaderboard report
Stock Market Prediction Astock RoBERTa WWM Ext (News) Recall 61.32 #12 of 17 Archive leaderboard report
Text Classification arXiv-10 RoBERTa Accuracy 0.779 #2 of 4 Archive leaderboard report
Type prediction ManyTypes4TypeScript RoBERTa Average Accuracy 59.84 #7 of 9 Archive leaderboard report
Type prediction ManyTypes4TypeScript RoBERTa Average F1 57.54 #7 of 9 Archive leaderboard report
Type prediction ManyTypes4TypeScript RoBERTa Average Precision 57.45 #7 of 9 Archive leaderboard report
Type prediction ManyTypes4TypeScript RoBERTa Average Recall 57.62 #7 of 9 Archive leaderboard report

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

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionRoBERTaSoftmaxWeight DecayWordPiece

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