{"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/challenges-in-data-to-document-generation","title":"Challenges in Data-to-Document Generation","arxiv_id":"1707.08052","date":"2017-07-25","proceeding":"EMNLP 2017 9","authors":["Sam Wiseman","Stuart M. Shieber","Alexander M. Rush"],"abstract":"Recent neural models have shown significant progress on the problem of\ngenerating short descriptive texts conditioned on a small number of database\nrecords. In this work, we suggest a slightly more difficult data-to-text\ngeneration task, and investigate how effective current approaches are on this\ntask. In particular, we introduce a new, large-scale corpus of data records\npaired with descriptive documents, propose a series of extractive evaluation\nmethods for analyzing performance, and obtain baseline results using current\nneural generation methods. Experiments show that these models produce fluent\ntext, but fail to convincingly approximate human-generated documents. Moreover,\neven templated baselines exceed the performance of these neural models on some\nmetrics, though copy- and reconstruction-based extensions lead to noticeable\nimprovements.","url_abs":"http://arxiv.org/abs/1707.08052v1","url_pdf":"http://arxiv.org/pdf/1707.08052v1.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":"challenges-in-data-to-document-generation","repo_url":"https://github.com/harvardnlp/boxscore-data","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"challenges-in-data-to-document-generation","repo_url":"https://github.com/harvardnlp/data2text","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"challenges-in-data-to-document-generation","repo_url":"https://github.com/KaijuML/rotowire-rg-metric","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"challenges-in-data-to-document-generation","repo_url":"https://github.com/ratishsp/data2text-1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"data-to-text-generation","task_name":"Data-to-Text Generation"},{"task_slug":"descriptive","task_name":"Descriptive"},{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[],"datasets_introduced":[{"slug":"rotowire","name":"RotoWire","full_name":"RotoWire"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/data-to-text-generation-on-rotowire","task":"Data-to-Text Generation","dataset":"RotoWire","model":"Encoder-decoder + conditional copy","rank_in_archive_order":6,"of":6,"metrics":{"BLEU":"14.19"},"uses_additional_data":false},{"leaderboard":"/sota/data-to-text-generation-on-rotowire-content","task":"Data-to-Text Generation","dataset":"RotoWire (Content Ordering)","model":"Encoder-decoder + conditional copy","rank_in_archive_order":5,"of":5,"metrics":{"BLEU":"14.49","DLD":"8.68%"},"uses_additional_data":false},{"leaderboard":"/sota/data-to-text-generation-on-rotowire-relation","task":"Data-to-Text Generation","dataset":"RotoWire (Relation Generation)","model":"Encoder-decoder + conditional copy","rank_in_archive_order":6,"of":6,"metrics":{"Precision":"74.80%","count":"23.72"},"uses_additional_data":false},{"leaderboard":"/sota/data-to-text-generation-on-rotowire-content-1","task":"Data-to-Text Generation","dataset":"Rotowire (Content Selection)","model":"Encoder-decoder + conditional copy","rank_in_archive_order":5,"of":5,"metrics":{"Precision":"29.49%","Recall":"36.18%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.08052","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}