Papers › BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation,...
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdel-rahman Mohamed, Omer Levy, Ves Stoyanov, Luke Zettlemoyer
We present BART, a denoising autoencoder for pretraining sequence-to-sequence models. BART is trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text. It uses a standard Tranformer-based neural machine translation architecture which, despite its simplicity, can be seen as generalizing BERT (due to the bidirectional encoder), GPT (with the left-to-right decoder), and many other more recent pretraining schemes. We evaluate a number of noising approaches, finding the best performance by both randomly shuffling the order of the original sentences and using a novel in-filling scheme, where spans of text are replaced with a single mask token. BART is particularly effective when fine tuned for text generation but also works well for comprehension tasks. It matches the performance of RoBERTa with comparable training resources on GLUE and SQuAD, achieves new state-of-the-art results on a range of abstractive dialogue, question answering, and summarization tasks, with gains of up to 6 ROUGE. BART also provides a 1.1 BLEU increase over a back-translation system for machine translation, with only target language pretraining. We also report ablation experiments that replicate other pretraining schemes within the BART framework, to better measure which factors most influence end-task performance.
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):
get_harvested_code_for_paper(arxiv_id="1910.13461")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="1910.13461", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="1910.13461")
Code
Syntology 53 code samples harvested from 9 repositories linked to this paper: 40 ran (of which 7 constructed an object rather than computing a result; 30 with no instrument failure: 1 honoured, 2 violated, 27 with no contract checked; 10 where Syntology's instrument failed) · 13 unverified. Where Syntology's instrument failed: 8 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.
By repository: community (archive-listed): 53 samples from 9 repositories, 40 ran (30 with no instrument failure; 10 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 8 of its repositories that Syntology harvested for other papers:
- allenai/scientific-claim-generation, community (archive-listed): harvested for another paper, listed under the repository
- asahi417/lm-question-generation, community (archive-listed): harvested for another paper, listed under the repository
- dawn0815/UniSA, community (archive-listed): harvested for another paper, listed under the repository
- huggingface/transformers, community (archive-listed): harvested for 4 other papers, listed under the repository
- jongwooko/nash-pruning-official, community (archive-listed): harvested for another paper, listed under the repository
- microsoft/Table-Pretraining, community (archive-listed): harvested for another paper, listed under the repository
- microsoft/fastseq, community (archive-listed): harvested for another paper, listed under the repository
- shijx12/kqapro_baselines, community (archive-listed): harvested for 2 other papers, listed under the repository
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.
47 repositories listed; official and paper-mentioned ones first.
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
53 samples harvested: 40 ran (of which 7 constructed an object rather than computing a result; 30 with no instrument failure: 1 honoured, 2 violated, 27 with no contract checked; 10 where Syntology's instrument failed) · 13 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.
Licence: 13 of the 53 samples are pointer only, meaning Syntology does not serve that copy's text. For 6 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 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”. For 1 sample that ran, the graph records a signature this page does not show: an argument's recorded type or element kind is a printout of an object in Syntology's harness (such as a mock object, a class or a pandas printout), not the type of an input, so it does not describe the call's shape. “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.
9b6e878d15f549bc · report
fixture: signature package_leaflets:list/1/PackageLeaflet · returned dict · level recorded 2026-09-24 6c663245fb8a9853 · report
fixture: signature package_leaflets_raw:list/1/Leaflet · returned int · level recorded 2026-09-24 e2635b61a3c6c441 · report
fixture: signature file_data:dict/0/? · returned tuple · level recorded 2026-09-24 5d81e0ebf38a1261 · report
fixture: signature doc_content:str/0/str, product_id:str/0/str · returned str · level recorded 2026-09-24 718d6d148c2907ad · report
fixture: signature package_leaflets:list/1/PackageLeaflet · returned list · level recorded 2026-09-24 e683aae8732f6e92 · report
fixture: signature package_leaflets_raw:list/1/Leaflet · returned list · level recorded 2026-09-24 cc1fd572c2800793 · report
fixture: signature string:str/0/str · returned str · level recorded 2026-09-24 13594c848a4f3848 · report
fixture: signature package_leaflets_Stanza:list/1/Leaflet · returned list · level recorded 2026-09-24 1d119e078734e42d · report
fixture: signature section_content:str/0/str · returned str · level recorded 2026-09-24 d0a738e3442f47f8 · report
fixture: signature package_leaflets:list/1/PackageLeaflet · returned list · level recorded 2026-09-24 5ab962d0a8a909a1 · report
fixture: signature package_leaflets:list/1/Leaflet · returned list · level recorded 2026-09-24 2d3ccf3606eee8d7 · report
03c622fbd053b445 · report
e3c82cdc2062e229 · report
e01d62d03dcd9ca4 · report
fixture: returned DefaultModule · level recorded 2026-09-22 8eadeed331b841fc · report
fixture: signature mask_token_id:int/0/int, masking_rate:float/0/float · returned Function · level recorded 2026-09-22 d43a426d78c2dacb · report
fixture: signature seq:list/1/list, num:int/0/int · returned list · level recorded 2026-09-22 548c3509c23676c4 · report
fixture: signature outputs:list/1/list, text_to_id:callable/0/?, marginalize:bool/0/bool · returned list · level recorded 2026-09-22 f05aeb96d74a93ac · report
a8821d279d5c47ec · report
20157e68a194d015 · report
390e7ee6a89b3db0 · report
fixture: signature sentenct:str/0/str, bos_token:str/0/str, eos_token:str/0/str · returned str · level recorded 2026-09-22 350f2fc5631039a2 · report
6b81c6ad0d553004 · report
fixture: signature prev_input_ids:list/1/list, bad_words_ids:list/1/list · returned list · level recorded 2026-09-25 8d77347cfa15b834 · report
fixture: signature prev_input_ids:Tensor/2/int, num_hypos:int/0/int, no_repeat_ngram_size:int/0/int, cur_len:int/0/int · returned list · level recorded 2026-09-24 2f97d78f96ebf790 · report
fixture: signature turn:dict/0/?, mm_state:dict/0/? · returned dict · level recorded 2026-09-24 d46c1839e7291c3f · report
fixture: signature turn:dict/0/? · returned list · level recorded 2026-09-24 8b437b06f0647a87 · report
fixture: signature stroke:str/0/str · returned list · level recorded 2026-09-24 05809fee3ec27f95 · report
fixture: signature config_path:str/0/str · returned dict · level recorded 2026-09-24 070b5bdeecd71825 · report
fixture: signature model_id:str/0/str, filename:str/0/str, use_cdn:bool/0/bool · returned str · level recorded 2026-09-24 874a42377fa0659f · report
fixture: signature attention_mask:Tensor/2/int · returned Tensor · level recorded 2026-09-22 899d1e2736236351 · report
fixture: signature url_or_filename:str/0/str · returned bool · level recorded 2026-09-24 52e211fe8bfc51fe · report
fixture: signature path:str/0/str · returned OrderedDict · level recorded 2026-09-24 f20c9b1aead27fea · report
fixture: signature input_ids:Tensor/2/int, padding_idx:int/0/int · returned Tensor · level recorded 2026-09-22 67aa808378a82feb · report
fixture: signature output_type:callable/0/?, config_class:str/0/str · returned function · level recorded 2026-09-25 census: 9 implementations from 11 papers, 3 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind int, dtype int64 src/model.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only ·
e00ccf7847d12b59 · report
fixture: signature input_ids:Tensor/2/int, pad_token_id:int/0/int · returned Tensor · level recorded 2026-09-22 census: 8 implementations from 8 papers, 4 distinct outputs across 2 buckets, one shared input per bucket src/generation_utils.py ran; the contract check could not run (RAISES) MIT (permissive) here; the stored copy is pointer only ·
0dc219ab6716a127 · report
fixture: signature logits:Tensor/2/float, top_k:int/0/int, top_p:float/0/float, filter_value:float/0/float, min_tokens_to_keep:int/0/int · returned Tensor · level recorded 2026-09-24 4ac94d84c40c32cb · report
fixture: signature var:Tensor/2/float · returned Tensor · level recorded 2026-09-22 e9a610ec98650659 · report
fixture: returned DataCollector · level recorded 2026-09-22 61295902b9488656 · report
fixture: returned ndarray · level recorded 2026-09-22 47a59a0b45c916c8 · report
fixture: signature length:int/0/int · returned str · level recorded 2026-09-22 425f7a153da6ce1c · report
fixture: returned BartParams · level recorded 2026-09-22 54be1be200dfeae7 · report
fixture: returned MemBartEncoderOutput · level recorded 2026-09-22 7dd81fe3cbe34bdb · report
fixture: returned MemBartSeq2SeqLMOutput · level recorded 2026-09-22 ff3f2f670826701f · report
fixture: returned MemBartSeq2SeqModelOutput · level recorded 2026-09-22 a38901f69351322f · report
fixture: returned UpdateGate · level recorded 2026-09-22 census: 13 implementations from 13 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32 memformers/models/membart/modeling_membart.py ran · our draft was wrong MIT (permissive) ·
ff8c2ca593924bd0 · report
fixture: signature mask:Tensor/2/float, dtype:dtype/0/?, tgt_len:int/0/int · returned Tensor · level recorded 2026-09-22 81708af0925e3ecb · report
a70369f5e1d4f10b · report
0f787d1f3f22c2bc · report
a47f8c473e362e5a · report
6af3a69713d1961b · 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.
7 assemblies on this paper: 7 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).
7 sibling assemblies for BART (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.
harvested Assembly for BART (graph method: BART). 2 bricks; individually on a fixture: 1 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; 1 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- generatorDataset entry · level 0 · unverified ·
350f2fc5631039a2 - add_to_special_tokens data_transform · level 1 · ran · our draft was wrong ·
390e7ee6a89b3db0
harvested Assembly for BART (graph method: BART). 10 bricks; individually on a fixture: 5 ran (of which 4 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) · 5 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- MemBartForConditionalGeneration entry · level 0 · unverified ·
a47f8c473e362e5a - MemBartModel model · level 0 · unverified ·
6af3a69713d1961b - MemBartSeq2SeqLMOutput helper · level 1 · ran ·
7dd81fe3cbe34bdb - MemBartEncoder model · level 0 · unverified ·
81708af0925e3ecb - MemBartEncoderOutput helper · level 1 · ran ·
54be1be200dfeae7 - MemBartSeq2SeqModelOutput helper · level 1 · ran ·
ff3f2f670826701f - MemBartEncoderLayer model · level 0 · unverified ·
0f787d1f3f22c2bc - UpdateGate helper · level 1 · ran ·
a38901f69351322f - _expand_mask helper · level 1 · ran · our draft was wrong ·
ff8c2ca593924bd0 - MemBartEncoderAttention model · level 0 · unverified ·
a70369f5e1d4f10b
harvested Assembly for BART (graph method: BART). 4 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) · 1 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- SumGen entry · level 0 · unverified ·
5a651b6ef42f5d35 - DataCollector helper · level 1 · ran ·
e9a610ec98650659 - auto_detach_to_cpu helper · level 1 · ran · our draft was wrong ·
61295902b9488656 - get_random_string helper · level 1 · ran · our draft was wrong ·
47a59a0b45c916c8
harvested Assembly for BART (graph method: BART). 1 brick; individually on a fixture: 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. The assembly has not been run end-to-end.
one of 7 siblings
- BartParams entry · level 1 · ran ·
425f7a153da6ce1c
harvested Assembly for BART (graph method: BART). 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 7 siblings
- DefaultModule entry · level 1 · ran ·
e01d62d03dcd9ca4 - LinearWarmupLR helper · level 1 · ran ·
0089735efb90328b
harvested Assembly for BART (graph method: BART). 1 brick; individually on a fixture: 1 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; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- text_infilling entry · level 1 · ran · our draft was wrong ·
8eadeed331b841fc
harvested Assembly for BART (graph method: BART). 5 bricks; individually on a fixture: 2 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; 2 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.
one of 7 siblings
- GENRE entry · level 0 · unverified ·
f05aeb96d74a93ac - GENREHubInterface helper · level 0 · unverified ·
a8821d279d5c47ec - _GENREHubInterface helper · level 0 · unverified ·
20157e68a194d015 - chunk_it helper · level 1 · ran; the contract check could not run (RAISES) ·
d43a426d78c2dacb - post_process_wikidata helper · level 1 · ran · our draft was wrong ·
548c3509c23676c4
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="1910.13461") returns these assemblies with each brick's role and wiring (how to connect).
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Abstractive Text Summarization | CNN / Daily Mail | BART | ROUGE-1 | 44.16 | #19 of 53 | Archive leaderboard | report |
| Abstractive Text Summarization | CNN / Daily Mail | BART | ROUGE-2 | 21.28 | #19 of 53 | Archive leaderboard | report |
| Abstractive Text Summarization | CNN / Daily Mail | BART | ROUGE-L | 40.90 | #19 of 53 | Archive leaderboard | report |
| Open-Domain Question Answering | ELI5 | BART | Rouge-1 | 30.6 | #3 of 6 | Archive leaderboard | report |
| Open-Domain Question Answering | ELI5 | BART | Rouge-2 | 6.2 | #3 of 6 | Archive leaderboard | report |
| Open-Domain Question Answering | ELI5 | BART | Rouge-L | 24.3 | #3 of 6 | Archive leaderboard | report |
| Question Answering | SQuAD1.1 dev | BART Base (with text infilling) | F1 | 90.8 | #52 of 55 | Archive leaderboard | report |
| Text Summarization | X-Sum | BART | ROUGE-1 | 45.14 | #7 of 18 | Archive leaderboard | report |
| Text Summarization | X-Sum | BART | ROUGE-2 | 22.27 | #7 of 18 | Archive leaderboard | report |
| Text Summarization | X-Sum | BART | ROUGE-3 | 37.25 | #7 of 18 | 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
Introduced by this paper: BART
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