CNN/DailyMail (lm-eval, See et al. v3.0.0)

lm-eval zero-shot abstractive summarization of CNN/DailyMail articles (See et al. version 3.0.0), scored with ROUGE-1/2/L and BERTScore.

Also known as: cnn_dailymail, CNN/DailyMail, CNN-DM, abisee/cnn_dailymail

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorygeneration
SubcategoryEnglish news abstractive summarization (non-anonymized CNN/DailyMail v3.0.0)
Page statusactive
MetricROUGE-1/2/L F-measure and BERTScore P/R/F1
Directionhigher_is_better
Dataset size11490
Dataset licenceApache-2.0 (Hub card; licensing section names version 1.0.0)
PublisherDeepMind (original collection); Stanford / Google Brain (See et al. non-anonymized summarization code)

What it measures

cnn_dailymail_abisee is the runnable lm-evaluation-harness task that loads Hugging Face abisee/cnn_dailymail config 3.0.0. The model reads a CNN or Daily Mail article and writes a short multi-sentence summary. References are the journalist-written highlight bullets. Version 3.0.0 is the non-anonymized summarization split associated with See, Liu, and Manning's pointer-generator work, not Hermann et al.'s original cloze reading-comprehension dump.

Task format

Zero-shot generation. Prompt: "Summarize the following article:" then the article then "Summary:". num_fewshot 0, max_gen_toks 128, do_sample false. YAML task field is cnn_dailymail_abisee; the directory is cnn_dailymail. The task README's "Tasks" list names cnn_dailymail, which does not match the YAML task key.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub