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 ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1902.00751")

Code

Syntology Ran 8 of 22 code samples harvested from 9 repositories linked to this paper; 14 have no recorded run. Of those that ran: 8 ran with no contract checked.

By repository: official repository: 1 sample from 1 repository, 0 ran; community (archive-listed): 21 samples from 8 repositories, 8 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

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

google-research/adapter-bert officialmentioned in papermentioned on GitHubtf report
Adapter-Hub/adapter-transformers mentioned on GitHubpytorchApache-2.0 report
AsaCooperStickland/Bert-n-Pals mentioned on GitHubpytorch report
CyberZHG/keras-bert mentioned on GitHubtfMIT report
Davidzhangyuanhan/NOAH mentioned on GitHubtf report
TATlong/keras_bert mentioned on GitHubtf report
ZhangYuanhan-AI/NOAH mentioned on GitHubtfMIT report
cs-mshah/Adapter-Bert mentioned on GitHubpytorch report
heekhero/DTL mentioned on GitHubpytorch report
hmohebbi/TF-Adapter-BERT mentioned on GitHubtf report
kpe/bert-for-tf2 mentioned on GitHubtfMIT report
osu-mlb/petl_vision mentioned on GitHubpytorch report
osu-mlb/vit_peft_vision mentioned on GitHubpytorch report
prrao87/fine-grained-sentiment mentioned on GitHubpytorch report
zphang/bert_on_stilts mentioned on GitHubpytorch 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.

Code Syntology ran Syntology

22 samples harvested; 8 ran; 0 honoured the contract we drafted; 14 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

8ran
14unverified

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; each sample names its own and says which. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

feedforward_adapter google-research/adapter-bert/modeling.py official repository unverified Apache-2.0 (permissive) · b42f7bbad5547ecf · report
Adapter osu-mlb/petl_vision/model/adapter.py community (archive-listed) ran no licence file found · pointer only · 7b012e4d16a33bf8 · report
Adapter zphang/bert_on_stilts/pytorch_pretrained_bert/modeling.py community (archive-listed) ran · metamorphic tier: invariant Apache-2.0 (permissive) · 69eca16a33f5afcb · report
AdapterSuper Davidzhangyuanhan/NOAH/model/module/adapter_super.py community (archive-listed) ran MIT (permissive) · 7a291b0c4ae38f59 · report
BERTLayerNorm AsaCooperStickland/Bert-n-Pals/modeling.py community (archive-listed) ran · metamorphic tier: invariant fingerprinted MIT (permissive) · fcde41b66078c9e5 · report
BERTLowRank AsaCooperStickland/Bert-n-Pals/modeling.py community (archive-listed) ran fingerprinted MIT (permissive) · 4fcb1d49ea7c6f51 · report
BERTSelfOutput AsaCooperStickland/Bert-n-Pals/modeling.py community (archive-listed) ran MIT (permissive) · c97326dd9a8f8a78 · report
Transformer prrao87/fine-grained-sentiment/training/transformer_utils/model.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 96aac0e3241b82e0 · report
TransformerWithAdapters prrao87/fine-grained-sentiment/training/transformer_utils/model.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 22acb9946a708133 · report
AdapterModule krypticmouse/Adapter-BERT/src/model/adapter.py community (archive-listed) unverified no licence file found · pointer only · cbac3bf75f2ee9c8 · report
AdapterModule cs-mshah/Adapter-Bert/adapter-bert/model/adapter.py community (archive-listed) unverified Apache-2.0 (permissive) · 2de542d0225f66f0 · report
build_model_from_config CyberZHG/keras-bert/keras_bert/loader.py community (archive-listed) unverified MIT (permissive) · 604dea8db99819dc · report
checkpoint_loader CyberZHG/keras-bert/keras_bert/loader.py community (archive-listed) unverified MIT (permissive) · 81518ffac57afb91 · report
extract_embeddings CyberZHG/keras-bert/keras_bert/util.py community (archive-listed) unverified MIT (permissive) · a8a537c8d7277f7e · report
extract_embeddings_generator CyberZHG/keras-bert/keras_bert/util.py community (archive-listed) unverified MIT (permissive) · 6afe8c057023f35c · report
gen_batch_inputs CyberZHG/keras-bert/keras_bert/bert.py community (archive-listed) unverified MIT (permissive) · 501f37049301636e · report
get_checkpoint_paths CyberZHG/keras-bert/keras_bert/util.py community (archive-listed) unverified MIT (permissive) · 13d48a14384b395e · report
get_inputs CyberZHG/keras-bert/keras_bert/layers/inputs.py community (archive-listed) unverified MIT (permissive) · 9cbbfb6d9b9dc99e · report
get_pretrained CyberZHG/keras-bert/keras_bert/datasets/pretrained.py community (archive-listed) unverified MIT (permissive) · 02f4590e932e79ef · report
get_token_embedding CyberZHG/keras-bert/keras_bert/bert.py community (archive-listed) unverified MIT (permissive) · 2276095f620fd668 · report
init_weight osu-mlb/petl_vision/model/adapter.py community (archive-listed) unverified no licence file found · pointer only · 332f932a85b80a1b · report
load_trained_model_from_checkpoint CyberZHG/keras-bert/keras_bert/loader.py community (archive-listed) unverified MIT (permissive) · 315e40f9424312ef · report

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