{"url":"/task/summarization","name":"Summarization","slug":"summarization","description_markdown":"Summarization is the task of producing a shorter version of one or several documents that preserves most of the\ninput's meaning.","categories":[{"name":"Natural Language Processing","url":"/area/natural-language-processing"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":8,"papers_with_code":4,"benchmarks":1,"benchmark_tables_in_archive":14,"benchmark_tables_shown":14,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":8,"subtasks":2,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/summarization-on-muld-vlsp","slug":"summarization-on-muld-vlsp","dataset":"MuLD (VLSP)","dataset_url":"/dataset/muld","rows_in_archive":2,"metrics":["BLEU-1","BLEU-4","METEOR","Rouge-L"],"first_row_in_archive_order":{"model":"Longformer","paper_title":"MuLD: The Multitask Long Document Benchmark","paper_url":"/paper/muld-the-multitask-long-document-benchmark","paper_date":"2022-02-15","arxiv_id":"2202.07362","code_links":[{"title":"ghomashudson/muld","url":"https://github.com/ghomashudson/muld"}],"syntology":null}},{"leaderboard":null,"slug":"summarization-on-bazzhangz-sumdataset","dataset":"bazzhangz/sumdataset","dataset_url":null,"rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-big-patent","dataset":"big_patent","dataset_url":"/dataset/bigpatent","rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-billsum","dataset":"BillSum","dataset_url":"/dataset/billsum","rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-ccdv-arxiv-summarization","dataset":"ccdv/arxiv-summarization","dataset_url":null,"rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-cnn-dailymail","dataset":"cnn_dailymail","dataset_url":"/dataset/cnn-daily-mail-1","rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-kmfoda-booksum","dataset":"kmfoda/booksum","dataset_url":null,"rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-launch-gov-report","dataset":"launch/gov_report","dataset_url":null,"rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-mlsum-de","dataset":"MLSUM de","dataset_url":"/dataset/mlsum","rows_in_archive":0,"metrics":["Rouge1"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-multi-news","dataset":"multi_news","dataset_url":"/dataset/multi-news","rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-samsum-1","dataset":"SAMSum","dataset_url":"/dataset/samsum-corpus","rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-samsum-corpus-a-human","dataset":"SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization","dataset_url":"/dataset/samsum-corpus","rows_in_archive":0,"metrics":["Test ROGUE-1","Test ROGUE-2","Test ROGUE-L","Validation ROGUE-1","Validation ROGUE-2","Validation ROGUE-L"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-scientific-papers","dataset":"scientific_papers","dataset_url":null,"rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss","meteor"],"first_row_in_archive_order":null},{"leaderboard":null,"slug":"summarization-on-xsum","dataset":"XSum","dataset_url":"/dataset/xsum","rows_in_archive":0,"metrics":["ROUGE-1","ROUGE-2","ROUGE-L","ROUGE-LSUM","gen_len","loss"],"first_row_in_archive_order":null}],"datasets":[{"url":"/dataset/cnn-daily-mail-1","name":"CNN/Daily Mail","full_name":"CNN/Daily Mail","num_papers_in_archive":530},{"url":"/dataset/samsum-corpus","name":"SAMSum","full_name":"","num_papers_in_archive":145},{"url":"/dataset/multi-news","name":"Multi-News","full_name":"Multi-News","num_papers_in_archive":122},{"url":"/dataset/bigpatent","name":"BigPatent","full_name":"","num_papers_in_archive":50},{"url":"/dataset/billsum","name":"BillSum","full_name":"BillSum","num_papers_in_archive":47},{"url":"/dataset/mlsum","name":"MLSUM","full_name":"MultiLingual SUMmarization","num_papers_in_archive":45},{"url":"/dataset/xsum","name":"XSum","full_name":null,"num_papers_in_archive":32},{"url":"/dataset/muld","name":"MuLD","full_name":"Multitask Long Document Benchmark","num_papers_in_archive":3}],"subtasks":[{"url":"/task/query-focused-summarization","name":"Query-focused Summarization"},{"url":"/task/unsupervised-extractive-summarization","name":"Unsupervised Extractive Summarization"}],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":4,"of":4,"tagged_in_all":8,"items":[{"url":"/paper/abstractive-text-summarization-using-sequence","title":"Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond","date":"2016-02-19","arxiv_id":"1602.06023","repositories_listed":4,"syntology":{"n":6,"n_ran":5,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/hierarchical-prompting-taxonomy-a-universal","title":"Hierarchical Prompting Taxonomy: A Universal Evaluation Framework for Large Language Models Aligned with Human Cognitive Principles","date":"2024-06-18","arxiv_id":"2406.12644","repositories_listed":1,"syntology":null},{"url":"/paper/muld-the-multitask-long-document-benchmark","title":"MuLD: The Multitask Long Document Benchmark","date":"2022-02-15","arxiv_id":"2202.07362","repositories_listed":1,"syntology":null},{"url":"/paper/sparsifying-transformer-models-with","title":"Sparsifying Transformer Models with Trainable Representation Pooling","date":"2020-09-10","arxiv_id":"2009.05169","repositories_listed":1,"syntology":null}],"syntology_records":1,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}