{"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/improving-abstraction-in-text-summarization","title":"Improving Abstraction in Text Summarization","arxiv_id":"1808.07913","date":"2018-08-23","proceeding":"EMNLP 2018 10","authors":["Wojciech Kryściński","Romain Paulus","Caiming Xiong","Richard Socher"],"abstract":"Abstractive text summarization aims to shorten long text documents into a\nhuman readable form that contains the most important facts from the original\ndocument. However, the level of actual abstraction as measured by novel phrases\nthat do not appear in the source document remains low in existing approaches.\nWe propose two techniques to improve the level of abstraction of generated\nsummaries. First, we decompose the decoder into a contextual network that\nretrieves relevant parts of the source document, and a pretrained language\nmodel that incorporates prior knowledge about language generation. Second, we\npropose a novelty metric that is optimized directly through policy learning to\nencourage the generation of novel phrases. Our model achieves results\ncomparable to state-of-the-art models, as determined by ROUGE scores and human\nevaluations, while achieving a significantly higher level of abstraction as\nmeasured by n-gram overlap with the source document.","url_abs":"http://arxiv.org/abs/1808.07913v1","url_pdf":"http://arxiv.org/pdf/1808.07913v1.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":[],"tasks":[{"task_slug":"abstractive-text-summarization","task_name":"Abstractive Text Summarization"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"text-summarization","task_name":"Text Summarization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/abstractive-text-summarization-on-cnn-daily","task":"Abstractive Text Summarization","dataset":"CNN / Daily Mail","model":"ML+RL ROUGE+Novel, with LM","rank_in_archive_order":44,"of":53,"metrics":{"ROUGE-1":"40.19","ROUGE-2":"17.38","ROUGE-L":"37.52"},"uses_additional_data":false},{"leaderboard":"/sota/text-summarization-on-cnn-daily-mail-2","task":"Text Summarization","dataset":"CNN / Daily Mail (Anonymized)","model":"ML+RL ROUGE+Novel, with LM","rank_in_archive_order":4,"of":13,"metrics":{"ROUGE-1":"40.02","ROUGE-2":"15.53","ROUGE-L":"37.44"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.07913","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}