Papers › Don't Stop Pretraining: Adapt Language Models to Domains and Tasks

Don't Stop Pretraining: Adapt Language Models to Domains and Tasks

23 Apr 2020ACL 2020 6arXiv:2004.10964archive 2025-07-28

Suchin Gururangan, Ana Marasović, Swabha Swayamdipta, Kyle Lo, Iz Beltagy, Doug Downey, Noah A. Smith

Language models pretrained on text from a wide variety of sources form the foundation of today's NLP. In light of the success of these broad-coverage models, we investigate whether it is still helpful to tailor a pretrained model to the domain of a target task. We present a study across four domains (biomedical and computer science publications, news, and reviews) and eight classification tasks, showing that a second phase of pretraining in-domain (domain-adaptive pretraining) leads to performance gains, under both high- and low-resource settings. Moreover, adapting to the task's unlabeled data (task-adaptive pretraining) improves performance even after domain-adaptive pretraining. Finally, we show that adapting to a task corpus augmented using simple data selection strategies is an effective alternative, especially when resources for domain-adaptive pretraining might be unavailable. Overall, we consistently find that multi-phase adaptive pretraining offers large gains in task performance.

PaperPDFConference PDFCodeCode 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="2004.10964")

Code

Syntology Ran 6 of 14 code samples harvested from 2 repositories linked to this paper; 8 have no recorded run. Of those that ran: 4 ran · our draft was wrong; 2 ran with no contract checked.

By repository: community (archive-listed): 13 samples from 2 repositories, 5 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

allenai/dont-stop-pretraining officialmentioned in paperpytorch report
SUSTechBruce/G-MAP mentioned on GitHubpytorch report
shizhediao/black-box-prompt-learning mentioned on GitHubpytorchApache-2.0 report
shizhediao/t-dna mentioned on GitHubpytorchApache-2.0 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

14 samples harvested; 6 ran; 0 honoured the contract we drafted; 8 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.

4ran · our draft was wrong
2ran
8unverified

Licence: 1 of the 14 samples is 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 2 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “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.

swish shizhediao/t-dna/TDNA/util.py community (archive-listed) ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 0f786c407fb1ee4c · report
gelu_new shizhediao/t-dna/TDNA/util.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · aef07bb1e2787431 · report
get_linear_schedule_with_warmup shizhediao/t-dna/TDNA/optimization.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 5bf56e40b4301199 · report
is_remote_url shizhediao/t-dna/TDNA/modeling.py community (archive-listed) ran Apache-2.0 (permissive) · 899d1e2736236351 · report
set_output_dir SUSTechBruce/G-MAP/domain_qa_ner/mem_roberta/run_finetune_qa_medication.py community (archive-listed) ran · our draft was wrong MIT (permissive) · bbc63fe66d3ee4d0 · report
default_compute_objective shizhediao/t-dna/TDNA/trainer.py community (archive-listed) unverified Apache-2.0 (permissive) · ada7a524687a5b73 · report
find_pruneable_heads_and_indices shizhediao/t-dna/TDNA/modeling.py community (archive-listed) unverified Apache-2.0 (permissive) · 1b4bc5e6ef0e9288 · report
get_constant_schedule shizhediao/t-dna/TDNA/optimization.py community (archive-listed) unverified Apache-2.0 (permissive) · 0cfd32e6d6d4419a · report
get_constant_schedule_with_warmup shizhediao/t-dna/TDNA/optimization.py community (archive-listed) unverified Apache-2.0 (permissive) · 2eeaa920a743dccc · report
get_logger shizhediao/t-dna/TDNA/util.py community (archive-listed) unverified Apache-2.0 (permissive) · 925a37d315d717fb · report
get_pairs shizhediao/t-dna/TDNA/tokenization.py community (archive-listed) unverified Apache-2.0 (permissive) · b50994ad24f6d583 · report
hf_bucket_url shizhediao/t-dna/TDNA/modeling.py community (archive-listed) unverified Apache-2.0 (permissive) · 3cbbdc1cef27d05d · report
nested_concat shizhediao/t-dna/TDNA/trainer.py community (archive-listed) unverified Apache-2.0 (permissive) · 8c6f5dccbea60d03 · report
to_list identical code first harvested elsewhere ran · our draft was wrong fingerprinted licence of this copy not recorded · 9df40357afea56cc · report

Tasks

Citation Intent Classification

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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