</> Code with Papers by Syntology

Datasets › CNN/Daily Mail

CNN/Daily Mail

Introduced by Ramesh Nallapati et al. in Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond16 Aug 2016 archive 2025-07-28

CNN/Daily Mail is a dataset for text summarization. Human generated abstractive summary bullets were generated from news stories in CNN and Daily Mail websites as questions (with one of the entities hidden), and stories as the corresponding passages from which the system is expected to answer the fill-in the-blank question. The authors released the scripts that crawl, extract and generate pairs of passages and questions from these websites.

In all, the corpus has 286,817 training pairs, 13,368 validation pairs and 11,487 test pairs, as defined by their scripts. The source documents in the training set have 766 words spanning 29.74 sentences on an average while the summaries consist of 53 words and 3.72 sentences.

Source: Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 8 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Abstractive Text Summarization CNN / Daily Mail Scrambled code + broken (alter) ROUGE-1 48.18 Universal Evasion Attacks on Summarization Scoring cestwc/universal-evasion 53 Compare
Document Summarization CNN / Daily Mail Scrambled code + broken (alter) ROUGE-1 48.18 Universal Evasion Attacks on Summarization Scoring cestwc/universal-evasion 26 Compare
Question Answering CNN / Daily Mail GA+MAGE (32) CNN 78.6 Linguistic Knowledge as Memory for Recurrent Neural Networks — 16 Compare
Extractive Text Summarization CNN / Daily Mail HAHSum ROUGE-1 44.68 Neural Extractive Summarization with Hierarchical... — 15 Compare
Text Summarization CNN / Daily Mail (Anonymized) HSSAS ROUGE-1 42.3 A Hierarchical Structured Self-Attentive Model for... — 13 Compare
Abstractive Text Summarization CNN/Daily Mail BART (TextBox 2.0) ROUGE-1 44.47 TextBox 2.0: A Text Generation Library with Pre-trained... RUCAIBox/TextBox 4 Compare
Text Generation CNN/Daily Mail PALM ROUGE-L 41.41 PALM: Pre-training an Autoencoding&Autoregressive... alibaba/AliceMind +1 1 Compare
Summarization cnn_dailymail no rows — — 0 Compare

Papers archive 2025-07-28

30 shown of 89 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 530. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Salient Information Prompting to Steer Content in Prompt-based Abstractive Summarization 1 1 3 Oct 2024 not harvested
CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generation 1 2 3 Oct 2024 not harvested
Segmented Recurrent Transformer: An Efficient Sequence-to-Sequence Model 1 1 24 May 2023 not harvested
Fourier Transformer: Fast Long Range Modeling by Removing Sequence Redundancy with FFT Operator 1 2 24 May 2023 not harvested
Align and Attend: Multimodal Summarization with Dual Contrastive Losses 2 1 13 Mar 2023 not harvested
TextBox 2.0: A Text Generation Library with Pre-trained Language Models 1 1 26 Dec 2022 ran 1 of 1 samples (0 unverified)
Universal Evasion Attacks on Summarization Scoring 1 3 25 Oct 2022 not harvested
Salience Allocation as Guidance for Abstractive Summarization 1 1 22 Oct 2022 not harvested
Calibrating Sequence likelihood Improves Conditional Language Generation 0 1 30 Sep 2022 not harvested
BRIO: Bringing Order to Abstractive Summarization 3 1 31 Mar 2022 ran 0 of 10 samples (10 unverified)
SummaReranker: A Multi-Task Mixture-of-Experts Re-ranking Framework for Abstractive Summarization 1 2 13 Mar 2022 ran 2 of 3 samples (1 unverified)
LongT5: Efficient Text-To-Text Transformer for Long Sequences 4 1 15 Dec 2021 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)
Learn to Copy from the Copying History: Correlational Copy Network for Abstractive Summarization 1 2 1 Nov 2021 not harvested
Considering Nested Tree Structure in Sentence Extractive Summarization with Pre-trained Transformer 0 1 1 Nov 2021 not harvested
Fastformer: Additive Attention Can Be All You Need 13 1 20 Aug 2021 ran 4 of 4 samples (0 unverified; 3 pointer-only for licence)
R-Drop: Regularized Dropout for Neural Networks 8 1 28 Jun 2021 ran 2 of 6 samples (4 unverified)
SimCLS: A Simple Framework for Contrastive Learning of Abstractive Summarization 2 1 3 Jun 2021 ran 4 of 6 samples (2 unverified; 1 pointer-only for licence)
IMPROVING ABSTRACTIVE SUMMARIZATION WITH SEGMENT-AUGMENTED AND POSITION-AWARENESS (ACLing2021) 1 1 1 Jun 2021 not harvested
Hie-BART: Document Summarization with Hierarchical BART 0 1 1 Jun 2021 not harvested
The Summary Loop: Learning to Write Abstractive Summaries Without Examples 1 1 11 May 2021 not harvested
Hierarchical Learning for Generation with Long Source Sequences 0 1 15 Apr 2021 not harvested
Mask Attention Networks: Rethinking and Strengthen Transformer 1 1 25 Mar 2021 ran 0 of 3 samples (3 unverified)
GLM: General Language Model Pretraining with Autoregressive Blank Infilling 8 2 18 Mar 2021 ran 1 of 1 samples (0 unverified)
Muppet: Massive Multi-task Representations with Pre-Finetuning 2 1 26 Jan 2021 not harvested
Subformer: A Parameter Reduced Transformer 0 1 1 Jan 2021 not harvested
Neural Extractive Summarization with Hierarchical Attentive Heterogeneous Graph Network 0 1 1 Nov 2020 not harvested
Better Fine-Tuning by Reducing Representational Collapse 3 1 6 Aug 2020 not harvested
Big Bird: Transformers for Longer Sequences 14 1 28 Jul 2020 ran 10 of 15 samples (5 unverified; 11 pointer-only for licence)
Synthesizer: Rethinking Self-Attention in Transformer Models 1 1 2 May 2020 ran 1 of 1 samples (0 unverified)
Extractive Summarization as Text Matching 2 3 19 Apr 2020 not harvested

The full list of 89 is in the JSON twin.

Dataset loaders archive 2025-07-28

huggingface.co/datasets/behind-the-cmo/behind-the-cmo-newsletterhuggingface/datasetstfpytorchjax
huggingface.co/datasets/cnn_dailymailhuggingface/datasetstfpytorchjax
huggingface.co/datasets/samyakmohelay/genai_datasethuggingface/datasetstfpytorchjax
huggingface.co/datasets/ZhongshengWang/Alpaca-cnn-dailymailhuggingface/datasetstfpytorchjax
huggingface.co/datasets/abisee/cnn_dailymailhuggingface/datasetstfpytorchjax
huggingface.co/datasets/texta1564/smart_sentryhuggingface/datasetstfpytorchjax
huggingface.co/datasets/ccdv/cnn_dailymailhuggingface/datasetstfpytorchjax
huggingface.co/datasets/malteos/test2huggingface/datasetstfpytorchjax
huggingface.co/datasets/nbroad/citesumhuggingface/datasetstfpytorchjax
huggingface.co/datasets/biglam/atypical_animacyhuggingface/datasetstfpytorchjax
huggingface.co/datasets/elihoole/asrs-aviation-reportshuggingface/datasetstfpytorchjax
huggingface.co/datasets/yhavinga/cnn_dailymail_dutchhuggingface/datasetstfpytorchjax
parl.ai/docs/tasks.html#cnn/dm-summarisationfacebookresearch/ParlAIpytorch
www.tensorflow.org/datasets/catalog/cnn_dailymailtensorflow/datasetstfjax
abisee/cnn-dailymailtf
docs.allennlp.org/models/main/models/generation/dataset_readers/cnn_dm/allenai/allennlp-modelspytorch

16 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

Question AnsweringText GenerationText SummarizationAbstractive Text SummarizationSequence-to-sequence Language ModelingDocument SummarizationSummarizationExtractive Text SummarizationExtractive Document Summarization

License archive 2025-07-28

MIT

Modalities archive 2025-07-28

Texts

Languages archive 2025-07-28

English

Variants archive 2025-07-28

  • ml6team/cnn_dailymail_nl
  • cnn_dailymail 3.0.0
  • yhavinga/cnn_dailymail_dutch
  • cnn_dailymail
  • CNN / Daily Mail (Anonymized)
  • CNN-DM
  • CNN / Daily Mail
  • CNN/Daily Mail

8 variant names, as the archive lists them.

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