metaai·lightalo unofficial · independent
Universe / Language & LLMs / BART
Language & LLMs · 2019

BART

Corrupt the text, learn to rebuild it — summarization's workhorse

open source superseded 140M (base) / 400M (large) params

Latest: BART base/large (Oct 2019)

Denoising sequence-to-sequence pretraining (October 2019): corrupt text with noise (deletion, infilling, shuffling), train a full encoder-decoder Transformer to reconstruct it. BART became the go-to model for summarization and generation tasks, and its recipe influenced a generation of seq2seq models.

Why it matters

BART's corrupt-and-reconstruct pretraining made a full encoder-decoder Transformer the go-to for summarization and generation; bart-large-cnn was the standard summarization baseline for years and still handles over a million monthly Hugging Face downloads. Its denoising recipe shaped the seq2seq models that followed it.

Facts

Try it yourself

Lineage

Descends fromfairseq

See the whole family tree →

Sources

More in Language & LLMs

RoBERTaXLM-RBlenderBotOPT-175BGalacticaLLaMA

Read the Language & LLMs story on the sky →

✦ Open on the map Explore Language & LLMs Quiz me