Papers › DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter

DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter

2 Oct 2019NeurIPS 2019 12arXiv:1910.01108archive 2025-07-28

Victor Sanh, Lysandre Debut, Julien Chaumond, Thomas Wolf

As Transfer Learning from large-scale pre-trained models becomes more prevalent in Natural Language Processing (NLP), operating these large models in on-the-edge and/or under constrained computational training or inference budgets remains challenging. In this work, we propose a method to pre-train a smaller general-purpose language representation model, called DistilBERT, which can then be fine-tuned with good performances on a wide range of tasks like its larger counterparts. While most prior work investigated the use of distillation for building task-specific models, we leverage knowledge distillation during the pre-training phase and show that it is possible to reduce the size of a BERT model by 40%, while retaining 97% of its language understanding capabilities and being 60% faster. To leverage the inductive biases learned by larger models during pre-training, we introduce a triple loss combining language modeling, distillation and cosine-distance losses. Our smaller, faster and lighter model is cheaper to pre-train and we demonstrate its capabilities for on-device computations in a proof-of-concept experiment and a comparative on-device study.

PaperPDFCodeCode 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 27 code samples harvested from 6 repositories linked to this paper: 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. Where Syntology's instrument failed: 6 where our contract draft was wrong; 3 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 1 sample from 1 repository, 1 ran (1 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 26 samples from 5 repositories, 20 ran (11 with no instrument failure; 9 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 7 of its repositories that Syntology harvested for other papers:

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.

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

huggingface/swift-coreml-transformers officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
huggingface/transformers officialmentioned in papermentioned 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
Karthik-Bhaskar/Context-Based-Question-Answering mentioned on GitHubtfApache-2.0Syntology: not harvested report
Milan-Chicago/GLG-Automated-Meta-data-Tagging mentioned on GitHubtfSyntology: not harvested report
allenai/scifact mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
askaydevs/distillbert-qa mentioned on GitHubpytorchSyntology: not harvested report
ayeffkay/rubert-tiny mentioned on GitHubpytorchSyntology: not harvested report
dngback/co-forget-protocol mentioned on GitHubApache-2.0Syntology: not harvested report
enzomuschik/distilfnd mentioned on GitHubpytorchMITSyntology: not harvested report
epfml/collaborative-attention mentioned on GitHubpytorchSyntology: 3 samples, 2 ran (1 instrument-failed) · 1 unverified report
facebookresearch/EgoTV mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
flexible-fl/flex-nlp mentioned on GitHubAGPL-3.0Syntology: not harvested report
frankaging/Causal-Distill mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Causal Distillation for Language Models (arXiv:2112.02505; 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) · 1 unverified); nothing checks that they implement this paper's method. report
franknb/Text-Summarization mentioned on GitHubSyntology: not harvested report
huggingface/node-question-answering mentioned on GitHubtfApache-2.0Syntology: not harvested report
huggingface/tflite-android-transformers mentioned on GitHubtfApache-2.0Syntology: not harvested report
jaketae/pytorch-malware-detection mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Longformer: The Long-Document Transformer (arXiv:2004.05150; that paper's own run record: 22 ran (of which 4 constructed an object rather than computing a result; 14 with no instrument failure: 1 honoured, 1 violated, 12 with no contract checked; 8 where Syntology's instrument failed) · 13 unverified); nothing checks that they implement this paper's method. report
knuddj1/op_text mentioned on GitHubpytorchMITSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
knuddy/op_text mentioned on GitHubpytorchMITSyntology: not harvested report
lukexyz/Deep-Lyrical-Genius mentioned on GitHubpytorchMITSyntology: not harvested report
mkavim/finetune_bert mentioned on GitHubtfSyntology: 14 samples, 11 ran (3 instrument-failed) · 3 unverified report
monologg/distilkobert mentioned on GitHubpytorchSyntology: 5 samples, 3 ran (2 instrument-failed) · 2 unverified report
msorkhpar/wiki-entity-summarization-preprocessor mentioned on GitHubpytorchCC-BY-4.0Syntology: not harvested report
nageshsinghc4/deepwrap mentioned on GitHubtfSyntology: 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
philschmid/knowledge-distillation-transformers-pytorch-sagemaker mentioned on GitHubpytorchMITSyntology: not harvested report
reycn/multi-modal-scale mentioned on GitHubpytorchMITSyntology: 3 samples, 3 ran (3 instrument-failed) · 0 unverified report
rybread1/stack-overflow-question-classification mentioned on GitHubnot reachable when probed 2026-09-17 — repositories for recent papers often appear after camera-readySyntology: not harvested report
sdadas/polish-roberta mentioned on GitHubpytorchLGPL-3.0Syntology: not harvested report
semantic-web-company/ptlm_wsid mentioned on GitHubSyntology: not harvested report
stefan-it/europeana-bert mentioned on GitHubtfMITSyntology: not harvested report
suinleelab/path_explain mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Axiomatic Attribution for Deep Networks (arXiv:1703.01365; that paper's own run record: 36 ran (of which 13 constructed an object rather than computing a result; 21 with no instrument failure: 3 honoured, 1 violated, 17 with no contract checked; 15 where Syntology's instrument failed) · 20 unverified); nothing checks that they implement this paper's method. report
tchebonenko/Automated-Topic_Modeling-and-NER mentioned on GitHubtfSyntology: not harvested report
twobooks/intro-aws-training mentioned on GitHubpytorchSyntology: not harvested report
2024-MindSpore-1/Code2 mindsporeSyntology: not harvested report
MindCode-4/code-3 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to MathOdyssey: Benchmarking Mathematical Problem-Solving Skills in Large Language Models... (arXiv:2406.18321; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
MindSpore-paper-code-2/code399 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

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

27 samples harvested: 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. “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.

3ran · honoured contract
1ran · violated contract
6ran · our draft was wrong
3ran; the contract check could not run
8ran
6unverified

Licence: 2 of the 27 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 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 6 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.

huggingface/swift-coreml-transformers official repository 1 sample, 1 ran (0 instrument-failed) · 0 unverified
timeit_and_report_mean model_generation/distilbert-onnx-coreml.py ran Apache-2.0 (permissive) · 3ae355fe5a2be50f · report fixture: signature f:callable/0/? · returned str · level recorded 2026-09-25
epfml/collaborative-attention community (archive-listed) 3 samples, 2 ran (1 instrument-failed) · 1 unverified
ArgsAdapterDistilBERT src/collaborative_attention/adapter_distilbert.py ran Apache-2.0 (permissive) · bd41263948c9f93a · report fixture: signature query:Tensor/3/float, key:Tensor/3/float, value:Tensor/3/float, mask:Tensor/2/float, head_mask:NoneType/0/? · returned Tensor · level recorded 2026-09-24
distilbert_layers src/collaborative_attention/adapter_distilbert.py ran · our draft was wrong Apache-2.0 (permissive) · 024e22b6643186b4 · report fixture: signature model:callable/0/? · returned ModuleList · level recorded 2026-09-22
DistilBERTCollaborativeAdapter src/collaborative_attention/adapter_distilbert.py unverified Apache-2.0 (permissive) · 299d28a1eb667c41 · report
knuddj1/op_text community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
get_confidence_scores op_text/utils.py ran fingerprinted MIT (permissive) · b07754f6f7d44e9e · report fixture: signature logits:Tensor/2/float · returned Tensor · level recorded 2026-09-24
mkavim/finetune_bert community (archive-listed) 14 samples, 11 ran (3 instrument-failed) · 3 unverified
DistillBertConfig finetune/modeling_distilbert.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · d3e9e3aeb78c5878 · report fixture: returned DistillBertConfig · level recorded 2026-09-22
DistillBertMultiHeadSelfAttention finetune/modeling_distilbert.py ran Apache-2.0 (permissive) · c73b35f8741d5bbc · report fixture: returned DistillBertMultiHeadSelfAttention · level recorded 2026-09-22
FeedForward in 198 papers finetune/modeling_distilbert.py ran Apache-2.0 (permissive) · 586463f8edb14bc7 · report fixture: returned FeedForward · level recorded 2026-09-22
PretrainedConfig finetune/modeling_distilbert.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · aa85de675f71b596 · report fixture: returned PretrainedConfig · level recorded 2026-09-22
create_position_ids finetune/modeling_distilbert.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 4e8314bcf966eece · report fixture: signature input_ids:EagerTensor/2/int · returned EagerTensor · level recorded 2026-09-22
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input
finetune/modeling_distilbert.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 9bc31d06256e1691 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
gelu_new in 55 papers
census: 3 implementations from 7 papers, 1 distinct output on one shared input
finetune/modeling_distilbert.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 5b3fd00218623502 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
get_initializer in 32 papers finetune/modeling_distilbert.py ran · our draft was wrong Apache-2.0 (permissive) · 333abb7d28cd1f9f · report fixture: signature initializer_range:float/0/float · returned TruncatedNormal · level recorded 2026-09-22
get_input_mask finetune/modeling_distilbert.py ran · violated contract fingerprinted Apache-2.0 (permissive) · 57e9478bd6df4915 · report fixture: signature input_ids:EagerTensor/2/int · returned EagerTensor · level recorded 2026-09-22
shape_list in 9 papers finetune/modeling_distilbert.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 6220c13c591c68a7 · report fixture: signature x:EagerTensor/3/float · returned list · level recorded 2026-09-22
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input
finetune/modeling_distilbert.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 86d13b9f67274f19 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
DistillBertModel finetune/modeling_distilbert.py unverified Apache-2.0 (permissive) · 1f3a00e205369e86 · report
DistillBertPretrained finetune/modeling_distilbert.py unverified Apache-2.0 (permissive) · 604d8efbff65993f · report
load_distillbert_model_weights_from_checkpoint finetune/modeling_distilbert.py unverified Apache-2.0 (permissive) · 4db4878c2df8732e · report
monologg/distilkobert community (archive-listed) 5 samples, 3 ran (2 instrument-failed) · 2 unverified
GroupedBatchSampler distillation/distiller.py ran Apache-2.0 (permissive) · c01d1e6d469d9bfd · report fixture: returned GroupedBatchSampler · level recorded 2026-09-22
_quantize distillation/distiller.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed Apache-2.0 (permissive) · f5d6d16633127806 · report fixture: signature x:list/1/float, bins:list/1/float · returned list · level recorded 2026-09-22
create_lengths_groups distillation/distiller.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed fingerprinted Apache-2.0 (permissive) · 2fe0a0a44bca4abd · report fixture: signature lengths:list/1/int, k:int/0/int · returned list · level recorded 2026-09-22
Distiller in 6 papers distillation/distiller.py unverified Apache-2.0 (permissive) · 20e3c004457d9336 · report
LmSeqsDataset distillation/distiller.py unverified Apache-2.0 (permissive) · 2efdc2800d06d37c · report
reycn/multi-modal-scale community (archive-listed) 3 samples, 3 ran (3 instrument-failed) · 0 unverified
build_transform in 170 papers script/3.evaluation.batch.py ran · our draft was wrong MIT (permissive) · 57e95166ade5f25c · report fixture: signature input_size:int/0/int · returned Compose · level recorded 2026-09-24
dynamic_preprocess in 40 papers script/3.evaluation.batch.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 6dc77bb22fb9d660 · report fixture: signature image:Image/0/?, min_num:int/0/int, max_num:int/0/int, image_size:int/0/int, use_thumbnail:bool/0/bool · returned list · level recorded 2026-09-22
find_closest_aspect_ratio in 45 papers script/3.evaluation.batch.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed MIT (permissive) here; the stored copy is pointer only · bd77f5f8067f18e9 · report fixture: signature aspect_ratio:float/0/float, target_ratios:list/1/tuple, width:int/0/int, height:int/0/int, image_size:int/0/int · returned tuple · 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 DistilBERT (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for DistilBERT (graph method: DistilBERT). 5 bricks; individually on a fixture: 3 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; 2 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. Distiller entry · level 0 · unverified · 20e3c004457d9336
  2. GroupedBatchSampler helper · level 1 · ran · c01d1e6d469d9bfd
  3. LmSeqsDataset helper · level 0 · unverified · 2efdc2800d06d37c
  4. create_lengths_groups helper · level 1 · ran; the contract check could not run (RAISES) · 2fe0a0a44bca4abd
  5. _quantize helper · level 1 · ran; the contract check could not run (RAISES) · f5d6d16633127806

harvested Assembly for DistilBERT (graph method: DistilBERT). 14 bricks; individually on a fixture: 11 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 3 honoured, 1 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. DistillBertModel entry · level 0 · unverified · 1f3a00e205369e86
  2. DistillBertConfig helper · level 1 · ran · d3e9e3aeb78c5878
  3. DistillBertMultiHeadSelfAttention model · level 1 · ran · c73b35f8741d5bbc
  4. DistillBertPretrained helper · level 0 · unverified · 604d8efbff65993f
  5. FeedForward model · level 1 · ran · 586463f8edb14bc7
  6. create_position_ids helper · level 1 · ran · our draft was wrong · 4e8314bcf966eece
  7. get_initializer helper · level 1 · ran · our draft was wrong · 333abb7d28cd1f9f
  8. get_input_mask helper · level 1 · ran · violated contract · 57e9478bd6df4915
  9. shape_list helper · level 1 · ran · our draft was wrong · 6220c13c591c68a7
  10. PretrainedConfig helper · level 1 · ran · aa85de675f71b596
  11. load_distillbert_model_weights_from_checkpoint helper · level 0 · unverified · 4db4878c2df8732e
  12. gelu helper · level 1 · ran · honoured contract · 9bc31d06256e1691
  13. gelu_new helper · level 1 · ran · honoured contract · 5b3fd00218623502
  14. swish helper · level 1 · ran · honoured contract · 86d13b9f67274f19

harvested Assembly for DistilBERT (graph method: DistilBERT). 4 bricks; individually on a fixture: 3 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; 1 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. DistilBERTCollaborativeAdapter entry · level 0 · unverified · 299d28a1eb667c41
  2. ArgsAdapterDistilBERT helper · level 1 · ran · bd41263948c9f93a
  3. CollaborativeLayerAdapter helper · level 1 · ran · e6c28dfaa7ebc812
  4. distilbert_layers helper · level 1 · ran · our draft was wrong · 024e22b6643186b4

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="1910.01108") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Hate Speech DetectionKnowledge DistillationLanguage ModelingLanguage ModellingLinguistic AcceptabilityNatural Language InferenceOnly Connect Walls Dataset Task 1 (Grouping)Question AnsweringSemantic Textual SimilaritySentiment AnalysisTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Linguistic Acceptability CoLA DistilBERT 66M Accuracy 49.1% #38 of 43 Archive leaderboard report
Natural Language Inference QNLI DistilBERT 66M Accuracy 90.2% #36 of 43 Archive leaderboard report
Natural Language Inference RTE DistilBERT 66M Accuracy 62.9% #68 of 90 Archive leaderboard report
Natural Language Inference WNLI DistilBERT 66M Accuracy 44.4 #23 of 23 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW DistilBERT (BASE) # Correct Groups 49 ± 4 #21 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW DistilBERT (BASE) # Solved Walls 0 ± 0 #21 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW DistilBERT (BASE) Adjusted Mutual Information (AMI) 14.0 ± .3 #21 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW DistilBERT (BASE) Adjusted Rand Index (ARI) 11.3 ± .3 #21 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW DistilBERT (BASE) Fowlkes Mallows Score (FMS) 29.1 ± .2 #21 of 22 Archive leaderboard report
Only Connect Walls Dataset Task 1 (Grouping) OCW DistilBERT (BASE) Wasserstein Distance (WD) 86.7 ± .6 #21 of 22 Archive leaderboard report
Question Answering MultiTQ DistillBERT Hits@1 8.3 #10 of 11 Archive leaderboard report
Question Answering MultiTQ DistillBERT Hits@10 48.4 #10 of 11 Archive leaderboard report
Question Answering Quora Question Pairs DistilBERT 66M Accuracy 89.2% #14 of 19 Archive leaderboard report
Question Answering SQuAD1.1 dev DistilBERT EM 77.7 #20 of 55 Archive leaderboard report
Question Answering SQuAD1.1 dev DistilBERT 66M F1 85.8 #54 of 55 Archive leaderboard report
Semantic Textual Similarity MRPC DistilBERT 66M Accuracy 90.2% #14 of 45 Archive leaderboard report
Semantic Textual Similarity STS Benchmark DistilBERT 66M Pearson Correlation 0.907 #16 of 66 Archive leaderboard report
Sentiment Analysis IMDb DistilBERT 66M Accuracy 92.82 #30 of 49 Archive leaderboard report
Sentiment Analysis SST-2 Binary classification DistilBERT 66M Accuracy 91.3 #55 of 87 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 ConnectionsDistilBERTDropoutKnowledge DistillationLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionSoftmaxWeight 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