Papers › Parameter-Efficient Transfer Learning for NLP

Parameter-Efficient Transfer Learning for NLP

2 Feb 2019arXiv:1902.00751archive 2025-07-28

Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin de Laroussilhe, Andrea Gesmundo, Mona Attariyan, Sylvain Gelly

Fine-tuning large pre-trained models is an effective transfer mechanism in NLP. However, in the presence of many downstream tasks, fine-tuning is parameter inefficient: an entire new model is required for every task. As an alternative, we propose transfer with adapter modules. Adapter modules yield a compact and extensible model; they add only a few trainable parameters per task, and new tasks can be added without revisiting previous ones. The parameters of the original network remain fixed, yielding a high degree of parameter sharing. To demonstrate adapter's effectiveness, we transfer the recently proposed BERT Transformer model to 26 diverse text classification tasks, including the GLUE benchmark. Adapters attain near state-of-the-art performance, whilst adding only a few parameters per task. On GLUE, we attain within 0.4% of the performance of full fine-tuning, adding only 3.6% parameters per task. By contrast, fine-tuning trains 100% of the parameters per task.

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 22 code samples harvested from 9 repositories linked to this paper: 14 ran (of which 3 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) · 8 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 21 samples from 8 repositories, 14 ran (14 with no instrument failure; 0 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 4 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.

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

google-research/adapter-bert officialmentioned in papermentioned on GitHubtfSyntology: 1 sample, 0 ran · 1 unverified report
Adapter-Hub/adapter-transformers mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
AsaCooperStickland/Bert-n-Pals mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
CyberZHG/keras-bert mentioned on GitHubtfMITSyntology: 10 samples, 6 ran (0 instrument-failed) · 4 unverified report
Davidzhangyuanhan/NOAH mentioned on GitHubtfSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
TATlong/keras_bert mentioned on GitHubtfSyntology: not harvested report
ZhangYuanhan-AI/NOAH mentioned on GitHubtfMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Neural Prompt Search (arXiv:2206.04673; that paper's own run record: 8 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 2 honoured, 0 violated, 2 with no contract checked; 4 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
cs-mshah/Adapter-Bert mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
heekhero/DTL mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Visual Prompt Tuning (arXiv:2203.12119; that paper's own run record: 18 ran (of which 12 constructed an object rather than computing a result; 16 with no instrument failure: 1 honoured, 1 violated, 14 with no contract checked; 2 where Syntology's instrument failed) · 9 unverified); nothing checks that they implement this paper's method. report
hmohebbi/TF-Adapter-BERT mentioned on GitHubtfSyntology: not harvested report
kpe/bert-for-tf2 mentioned on GitHubtfMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ALBERT: A Lite BERT for Self-supervised Learning of Language Representations (arXiv:1909.11942; that paper's own run record: 81 ran (of which 17 constructed an object rather than computing a result; 56 with no instrument failure: 4 honoured, 0 violated, 52 with no contract checked; 25 where Syntology's instrument failed) · 45 unverified); nothing checks that they implement this paper's method. report
osu-mlb/petl_vision mentioned on GitHubpytorchSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
osu-mlb/vit_peft_vision mentioned on GitHubpytorchSyntology: not harvested report
prrao87/fine-grained-sentiment mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
zphang/bert_on_stilts mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
krypticmouse/Adapter-BERT pytorchSyntology: 1 sample, 0 ran · 1 unverified report
mindspore-courses/d2l-mindspore mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SqueezeBERT: What can computer vision teach NLP about efficient neural networks? (arXiv:2006.11316; 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

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

22 samples harvested: 14 ran (of which 3 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) · 8 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

14ran
8unverified

Licence: 3 of the 22 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 9 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.

google-research/adapter-bert official repository 1 sample, 0 ran · 1 unverified
feedforward_adapter modeling.py unverified Apache-2.0 (permissive) · b42f7bbad5547ecf · report
AsaCooperStickland/Bert-n-Pals community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
BERTLayerNorm modeling.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · fcde41b66078c9e5 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
BERTLowRank modeling.py ran fingerprinted MIT (permissive) · 4fcb1d49ea7c6f51 · report fixture: signature hidden_states:Tensor/3/float, attention_mask:NoneType/0/? · returned Tensor · level recorded 2026-09-24
BERTSelfOutput modeling.py ran MIT (permissive) · c97326dd9a8f8a78 · report fixture: signature hidden_states:Tensor/2/float, input_tensor:Tensor/2/float, attention_mask:NoneType/0/?, i:int/0/int · returned Tensor · level recorded 2026-09-24
CyberZHG/keras-bert community (archive-listed) 10 samples, 6 ran (0 instrument-failed) · 4 unverified
checkpoint_loader keras_bert/loader.py ran MIT (permissive) · 81518ffac57afb91 · report fixture: signature checkpoint_file:str/0/str · returned function · level recorded 2026-09-25
extract_embeddings_generator keras_bert/util.py ran MIT (permissive) · 6afe8c057023f35c · report fixture: signature model:str/0/str, texts:list/1/str, poolings:NoneType/0/?, vocabs:dict/0/?, cased:bool/0/bool, batch_size:int/0/int, cut_embed:bool/0/bool, output_layer_num:int/0/int · returned generator · level recorded 2026-09-25
gen_batch_inputs keras_bert/bert.py ran MIT (permissive) · 501f37049301636e · report fixture: signature sentence_pairs:list/1/list, token_dict:dict/0/?, token_list:list/1/str, seq_len:int/0/int, mask_rate:float/0/float, mask_mask_rate:float/0/float, mask_random_rate:float/0/float, swap_sentence_rate:float/0/float, force_mask:bool/0/bool · returned tuple · level recorded 2026-09-25
get_checkpoint_paths keras_bert/util.py ran MIT (permissive) · 13d48a14384b395e · report fixture: signature model_path:str/0/str · returned CheckpointPaths · level recorded 2026-09-25
get_inputs in 14 papers keras_bert/layers/inputs.py ran MIT (permissive) · 9cbbfb6d9b9dc99e · report fixture: signature seq_len:int/0/int · returned list · level recorded 2026-09-25
get_token_embedding keras_bert/bert.py ran MIT (permissive) · 2276095f620fd668 · report fixture: signature model:callable/0/? · returned Parameter · level recorded 2026-09-25
build_model_from_config keras_bert/loader.py unverified MIT (permissive) · 604dea8db99819dc · report
extract_embeddings in 9 papers keras_bert/util.py unverified MIT (permissive) · a8a537c8d7277f7e · report
get_pretrained keras_bert/datasets/pretrained.py unverified MIT (permissive) · 02f4590e932e79ef · report
load_trained_model_from_checkpoint keras_bert/loader.py unverified MIT (permissive) · 315e40f9424312ef · report
Davidzhangyuanhan/NOAH community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
AdapterSuper model/module/adapter_super.py ran MIT (permissive) · 7a291b0c4ae38f59 · report fixture: returned AdapterSuper · level recorded 2026-09-22
cs-mshah/Adapter-Bert community (archive-listed) 1 sample, 0 ran · 1 unverified
AdapterModule adapter-bert/model/adapter.py unverified Apache-2.0 (permissive) · 2de542d0225f66f0 · report
osu-mlb/petl_vision community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
Adapter in 28 papers model/adapter.py ran no licence file found · pointer only · 7b012e4d16a33bf8 · report fixture: signature x:Tensor/2/float, add_residual:bool/0/bool, residual:NoneType/0/? · returned Tensor · level recorded 2026-09-24
init_weight in 5 papers model/adapter.py unverified no licence file found · pointer only · 332f932a85b80a1b · report
prrao87/fine-grained-sentiment community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
Transformer in 173 papers training/transformer_utils/model.py ran · metamorphic tier: deterministic MIT (permissive) · 96aac0e3241b82e0 · report fixture: returned Transformer · level recorded 2026-09-22
TransformerWithAdapters training/transformer_utils/model.py ran · metamorphic tier: deterministic MIT (permissive) · 22acb9946a708133 · report fixture: returned TransformerWithAdapters · level recorded 2026-09-22
zphang/bert_on_stilts community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
Adapter in 28 papers pytorch_pretrained_bert/modeling.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 69eca16a33f5afcb · report fixture: signature hidden_states:Tensor/3/float · returned Tensor · level recorded 2026-09-22
krypticmouse/Adapter-BERT community (archive-listed) 1 sample, 0 ran · 1 unverified
AdapterModule src/model/adapter.py unverified no licence file found · pointer only · cbac3bf75f2ee9c8 · 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.

8 assemblies on this paper: 8 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

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

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

one of 8 siblings

  1. AdapterModule entry · level 0 · unverified · cbac3bf75f2ee9c8

harvested Assembly for transfer with adapter modules (graph method: transfer with adapter modules). 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 8 siblings

  1. TransformerWithAdapters entry · level 1 · ran · 22acb9946a708133
  2. Transformer model · level 1 · ran · 96aac0e3241b82e0

harvested Assembly for transfer with adapter modules (graph method: transfer with adapter modules). 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 8 siblings

  1. AdapterSuper entry · level 1 · ran · 7a291b0c4ae38f59
  2. QuickGELU helper · level 1 · ran · 990ae00bb84c8cce

harvested Assembly for transfer with adapter modules (graph method: transfer with adapter modules). 2 bricks; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 8 siblings

  1. feedforward_adapter entry · level 0 · unverified · b42f7bbad5547ecf
  2. gelu helper · level 1 · ran · honoured contract · d08f324f950148de

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

one of 8 siblings

  1. AdapterModule entry · level 0 · unverified · 2de542d0225f66f0

harvested Assembly for transfer with adapter modules (graph method: transfer with adapter modules). 4 bricks; individually on a fixture: 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 8 siblings

  1. BERTSelfOutput entry · level 1 · ran · c97326dd9a8f8a78
  2. BERTLayerNorm helper · level 1 · ran · fcde41b66078c9e5
  3. BERTLowRank model · level 1 · ran · 4fcb1d49ea7c6f51
  4. gelu helper · level 1 · ran · honoured contract · 40e9fee2e0b7e278

harvested Assembly for transfer with adapter modules (graph method: transfer with adapter modules). 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 8 siblings

  1. Adapter entry · level 1 · ran · 7b012e4d16a33bf8
  2. init_weight helper · level 0 · unverified · 332f932a85b80a1b

harvested Assembly for transfer with adapter modules (graph method: transfer with adapter modules). 3 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 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 8 siblings

  1. Adapter entry · level 1 · ran · 69eca16a33f5afcb
  2. gelu helper · level 1 · ran · honoured contract · fdc64f4c72036ae4
  3. swish helper · level 2 · ran · our draft was wrong · 0f786c407fb1ee4c

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="1902.00751") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image ClassificationText ClassificationTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification OmniBenchmark Adapter-ViTB/16 Average Top-1 Accuracy 44.5 #4 of 22 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

Absolute Position EncodingsAdamAttentionAttention DropoutBERTBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSoftmaxTransformerWeight 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