BART

E435868

denoising autoencoder neural network architecture sequence-to-sequence transformer model

BART is a sequence-to-sequence transformer model developed by Facebook AI for tasks like text generation, summarization, and translation.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (1)

Label	Occurrences
BART canonical	1

Statements (52)

Predicate	Object
instanceOf	denoising autoencoder ⓘ neural network architecture ⓘ sequence-to-sequence transformer model ⓘ
applicationDomain	natural language processing ⓘ
architectureType	encoder-decoder ⓘ
basedOn	Transformer architecture ⓘ
combinesIdeasFrom	BERT NERFINISHED ⓘ GPT NERFINISHED ⓘ
developer	FAIR NERFINISHED ⓘ Facebook AI NERFINISHED ⓘ Facebook AI Research NERFINISHED ⓘ
hasComponent	decoder ⓘ encoder ⓘ
hasVariant	BART-base NERFINISHED ⓘ BART-large NERFINISHED ⓘ BART-large-CNN NERFINISHED ⓘ BART-large-XSum NERFINISHED ⓘ MBART NERFINISHED ⓘ
inputType	text ⓘ
introducedInPaper	BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension NERFINISHED ⓘ
introducedYear	2019 ⓘ
language	English ⓘ
license	MIT-like (via Fairseq, depending on distribution) ⓘ
openSourceImplementation	Fairseq NERFINISHED ⓘ Hugging Face Transformers NERFINISHED ⓘ
optimizationAlgorithm	Adam NERFINISHED ⓘ
outputType	text ⓘ
paperAuthors	Abdelrahman Mohamed NERFINISHED ⓘ Luke Zettlemoyer NERFINISHED ⓘ Marjan Ghazvininejad NERFINISHED ⓘ Mike Lewis NERFINISHED ⓘ Naman Goyal NERFINISHED ⓘ Omer Levy NERFINISHED ⓘ Veselin Stoyanov NERFINISHED ⓘ Yinhan Liu NERFINISHED ⓘ
pretrained	true ⓘ
pretrainingStrategy	corrupt-then-reconstruct ⓘ
releasedBy	Facebook AI NERFINISHED ⓘ
supportsTask	abstractive summarization ⓘ dialogue generation ⓘ machine translation ⓘ sequence tagging ⓘ text classification ⓘ text generation ⓘ
trainingObjective	denoising autoencoding ⓘ sequence-to-sequence language modeling ⓘ
usesNoiseType	document rotation ⓘ sentence permutation ⓘ text infilling ⓘ token deletion ⓘ token masking ⓘ
usesSubwordTokenization	true ⓘ

How these facts were elicited

The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.

Instruction

You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10.

# Requirements
- If you don't know the subject at all, return an empty list.
- If the subject is not a named entity, return an empty list.
- Include at least one triple where predicate is "instanceOf".
- Do not get too wordy.
- Separate several objects into multiple triples with one object.

Input

Subject: BART
Description of subject: BART is a sequence-to-sequence transformer model developed by Facebook AI for tasks like text generation, summarization, and translation.

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Hugging Face Transformers → supportsModelType → BART ⓘ