{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/howsumm-a-multi-document-summarization","title":"HowSumm: A Multi-Document Summarization Dataset Derived from WikiHow Articles","arxiv_id":"2110.03179","date":"2021-10-07","proceeding":null,"authors":["Odellia Boni","Guy Feigenblat","Guy Lev","Michal Shmueli-Scheuer","Benjamin Sznajder","David Konopnicki"],"abstract":"We present HowSumm, a novel large-scale dataset for the task of query-focused multi-document summarization (qMDS), which targets the use-case of generating actionable instructions from a set of sources. This use-case is different from the use-cases covered in existing multi-document summarization (MDS) datasets and is applicable to educational and industrial scenarios. We employed automatic methods, and leveraged statistics from existing human-crafted qMDS datasets, to create HowSumm from wikiHow website articles and the sources they cite. We describe the creation of the dataset and discuss the unique features that distinguish it from other summarization corpora. Automatic and human evaluations of both extractive and abstractive summarization models on the dataset reveal that there is room for improvement.","url_abs":"https://arxiv.org/abs/2110.03179v2","url_pdf":"https://arxiv.org/pdf/2110.03179v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"howsumm-a-multi-document-summarization","repo_url":"https://github.com/odelliab/HowSumm","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"abstractive-text-summarization","task_name":"Abstractive Text Summarization"},{"task_slug":"articles","task_name":"Articles"},{"task_slug":"document-summarization","task_name":"Document Summarization"},{"task_slug":"multi-document-summarization","task_name":"Multi-Document Summarization"}],"methods":[],"datasets_introduced":[{"slug":"howsumm","name":"HowSumm","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"LexRank (query: method + article + steps titles)","rank_in_archive_order":1,"of":9,"metrics":{"ROUGE-1":"53.5"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"CES (query: method + article + steps titles)","rank_in_archive_order":2,"of":9,"metrics":{"ROUGE-1":"52.2"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"GreedyRel (query: method + article + steps titles)","rank_in_archive_order":3,"of":9,"metrics":{"ROUGE-1":"48.6"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"CES (query: method title)","rank_in_archive_order":4,"of":9,"metrics":{"ROUGE-1":"48.4"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"CES (query: method + article titles)","rank_in_archive_order":5,"of":9,"metrics":{"ROUGE-1":"48.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"LexRank (query: method title)","rank_in_archive_order":6,"of":9,"metrics":{"ROUGE-1":"47.7"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"LexRank (query: method + article titles)","rank_in_archive_order":7,"of":9,"metrics":{"ROUGE-1":"47.1"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"GreedyRel (query: method title)","rank_in_archive_order":8,"of":9,"metrics":{"ROUGE-1":"43.4"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-method","task":"Document Summarization","dataset":"HowSumm-Method","model":"GreedyRel (query: method + article titles)","rank_in_archive_order":9,"of":9,"metrics":{"ROUGE-1":"42.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"LexRank (query: step title)","rank_in_archive_order":1,"of":11,"metrics":{"ROUGE-1":"39.6"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"CES (query: step title)","rank_in_archive_order":2,"of":11,"metrics":{"ROUGE-1":"39.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"CES (query: step + method titles)","rank_in_archive_order":3,"of":11,"metrics":{"ROUGE-1":"38.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"LexRank (query: step + method titles)","rank_in_archive_order":4,"of":11,"metrics":{"ROUGE-1":"38.2"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"CES (query: step + method + article titles)","rank_in_archive_order":5,"of":11,"metrics":{"ROUGE-1":"37.0"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"LexRank (query: step + method + article titles)","rank_in_archive_order":6,"of":11,"metrics":{"ROUGE-1":"36.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"GreedyRel (query: step + method titles)","rank_in_archive_order":7,"of":11,"metrics":{"ROUGE-1":"30.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"GreedyRel (query: step title)","rank_in_archive_order":8,"of":11,"metrics":{"ROUGE-1":"30.1"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"BM25-HierSumm (query: step + method titles)","rank_in_archive_order":9,"of":11,"metrics":{"ROUGE-1":"23.0"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"BM25-HierSumm (query: step title)","rank_in_archive_order":10,"of":11,"metrics":{"ROUGE-1":"22.3"},"uses_additional_data":false},{"leaderboard":"/sota/document-summarization-on-howsumm-step","task":"Document Summarization","dataset":"HowSumm-Step","model":"BM25-HierSumm (query: step + method + article titles)","rank_in_archive_order":11,"of":11,"metrics":{"ROUGE-1":"21.9"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2110.03179","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}