Methods › Sequential › Sequence To Sequence Models › BART

BART

1,642 papers tagged archive 2025-07-28

Introduced by Mike Lewis et al. in BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

BART is a denoising autoencoder for pretraining sequence-to-sequence models. It 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 Transformer-based neural machine translation architecture. It uses a standard seq2seq/NMT architecture with a bidirectional encoder (like BERT) and a left-to-right decoder (like GPT). This means the encoder's attention mask is fully visible, like BERT, and the decoder's attention mask is causal, like GPT2.

PaperSource

Papers archive 2025-07-28

30 shown of 1,642, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 485 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
RAG1,278
Retrieval-augmented Generation1,081
Retrieval950
Question Answering347
Language Modelling190
Language Modeling158
Large Language Model129
Hallucination119
Text Generation111
Information Retrieval107
Knowledge Graphs59
Sentence56
Benchmarking54
Abstractive Text Summarization50
Decoder50
Response Generation46
Answer Generation40
Reranking36
Text Summarization36
Articles35

Usage over time archive 2025-07-28

Papers per year tagged with BART: 2019 to 2025, peak 735 735 0 2019: 3 papers 2019 2020: 37 papers 2020 2021: 94 papers 2021 2022: 93 papers 2022 2023: 103 papers 2023 2024: 735 papers 2024 2025: 577 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (1,642 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Sequence To Sequence ModelsTransformers

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