language:
- en
tags:
- summarization
license: mit
thumbnail: https://huggingface.co/front/thumbnails/facebook.png
model-index:
- name: facebook/bart-large-cnn
results:
- task:
type: summarization
name: Summarization
dataset:
name: cnn_dailymail
type: cnn_dailymail
config: 3.0.0
split: train
metrics:
- name: ROUGE-1
type: rouge
value: 42.9486
verified: true
- name: ROUGE-2
type: rouge
value: 20.8149
verified: true
- name: ROUGE-L
type: rouge
value: 30.6186
verified: true
- name: ROUGE-LSUM
type: rouge
value: 40.0376
verified: true
- name: loss
type: loss
value: 2.529000997543335
verified: true
- name: gen_len
type: gen_len
value: 78.5866
verified: true
BART (large-sized model), fine-tuned on CNN Daily Mail
BART model pre-trained on English language, and fine-tuned on CNN Daily Mail. It was introduced in the paper BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension by Lewis et al. and first released in [this repository (https://github.com/pytorch/fairseq/tree/master/examples/bart).
Disclaimer: The team releasing BART did not write a model card for this model so this model card has been written by the Hugging Face team.
Model description
BART is a transformer encoder-encoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder. BART is pre-trained by (1) corrupting text with an arbitrary noising function, and (2) learning a model to reconstruct the original text.
BART is particularly effective when fine-tuned for text generation (e.g. summarization, translation) but also works well for comprehension tasks (e.g. text classification, question answering). This particular checkpoint has been fine-tuned on CNN Daily Mail, a large collection of text-summary pairs.
Intended uses & limitations
You can use this model for text summarization.
How to use
Here is how to use this model with the pipeline API:
from transformers import pipeline
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
ARTICLE = """ from transformers import pipeline
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
ARTICLE = """ Professor Michael Adams is a specialist in Australian corporate law and international corporate governance. Michael has expertise in financial services regulation, information governance, consumer protection and the broader area of legal technology. Professor Adams is the Academic Dean of the University of New England Law School and from 2007 to 2017 Dean of Law at Western Sydney Law School.
He teaches, writes and presents regularly on these topics, as well as a major media commentator. He has published 12 books, 37 chapters, >120 articles and presented at >300 conferences/seminars in the last 30 years.
Professor Adams’ has degrees in accounting/economics and law, with postgraduate studies at University College London; he is a Fellow of the Australian College of Educators; the Governance Institute of Australia [FGA/FGIA(Life)] and the Australian Academy of Law; Member of Resolution Institute. He was the UTS Professor of Corporate Law and Perpetual Trustees Australia chair of Financial Services Law.
2001 National President of Governance Institute of Australia and served on the ICSA International Professional Standards Committee. He has been a consultant to national law firm Ashurst and currently, Coleman Grieg Lawyers. Director of GIA, charities Freedom Hub Ltd, ACOM theological college; and the Advisory Board of Information Governance ANZ; previously director of the Australian Pro Bono Centre; and the Australian Academy of Law. Former President of the Australasian Law Academics Association; the Society of Corporate Law Academics and deputy chair of the Australian Council of Law Deans.
In 2020 named Academic Lawyer of the Year by Lawyers Weekly in the 20th Australian Law Awards. 2000 Australian University Teacher of the Year in “Law and Legal Studies” and in 2005, recipient of the Governance Institute President’s Award and later made a Life member of both ALTA and the Governance Institute.
"""
print(summarizer(ARTICLE, max_length=130, min_length=30, do_sample=False))
>>> [{'summary_text': 'Liana Barrientos, 39, is charged with two counts of "offering a false instrument for filing in the first degree" In total, she has been married 10 times, with nine of her marriages occurring between 1999 and 2002. She is believed to still be married to four men.'}]
BibTeX entry and citation info
@article{DBLP:journals/corr/abs-1910-13461,
author = {Mike Lewis and
Yinhan Liu and
Naman Goyal and
Marjan Ghazvininejad and
Abdelrahman Mohamed and
Omer Levy and
Veselin Stoyanov and
Luke Zettlemoyer},
title = {{BART:} Denoising Sequence-to-Sequence Pre-training for Natural Language
Generation, Translation, and Comprehension},
journal = {CoRR},
volume = {abs/1910.13461},
year = {2019},
url = {http://arxiv.org/abs/1910.13461},
eprinttype = {arXiv},
eprint = {1910.13461},
timestamp = {Thu, 31 Oct 2019 14:02:26 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1910-13461.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}