{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T04:45:36Z","timestamp":1777092336752,"version":"3.51.4"},"reference-count":50,"publisher":"MIT Press - Journals","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Transactions of the Association for Computational Linguistics"],"published-print":{"date-parts":[[2019,11]]},"abstract":"<jats:p> We present DREAM, the first dialogue-based multiple-choice reading comprehension data set. Collected from English as a Foreign Language examinations designed by human experts to evaluate the comprehension level of Chinese learners of English, our data set contains 10,197 multiple-choice questions for 6,444 dialogues. In contrast to existing reading comprehension data sets, DREAM is the first to focus on in-depth multi-turn multi-party dialogue understanding. DREAM is likely to present significant challenges for existing reading comprehension systems: 84% of answers are non-extractive, 85% of questions require reasoning beyond a single sentence, and 34% of questions also involve commonsense knowledge. <\/jats:p><jats:p> We apply several popular neural reading comprehension models that primarily exploit surface information within the text and find them to, at best, just barely outperform a rule-based approach. We next investigate the effects of incorporating dialogue structure and different kinds of general world knowledge into both rule-based and (neural and non-neural) machine learning-based reading comprehension models. Experimental results on the DREAM data set show the effectiveness of dialogue structure and general world knowledge. DREAM is available at https:\/\/dataset.org\/dream\/ . <\/jats:p>","DOI":"10.1162\/tacl_a_00264","type":"journal-article","created":{"date-parts":[[2019,4,29]],"date-time":"2019-04-29T18:09:19Z","timestamp":1556561359000},"page":"217-231","source":"Crossref","is-referenced-by-count":91,"title":["DREAM: A Challenge Data Set and Models for Dialogue-Based Reading Comprehension"],"prefix":"10.1162","volume":"7","author":[{"given":"Kai","family":"Sun","sequence":"first","affiliation":[{"name":"Cornell University, Ithaca, NY, USA."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dian","family":"Yu","sequence":"additional","affiliation":[{"name":"Tencent AI Lab, Bellevue, WA, USA."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianshu","family":"Chen","sequence":"additional","affiliation":[{"name":"Tencent AI Lab, Bellevue, WA, USA."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dong","family":"Yu","sequence":"additional","affiliation":[{"name":"Tencent AI Lab, Bellevue, WA, USA."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yejin","family":"Choi","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, WA, USA"},{"name":"Allen Institute for Artificial Intelligence, Seattle, WA, USA."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Claire","family":"Cardie","sequence":"additional","affiliation":[{"name":"Cornell University, Ithaca, NY, USA."}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","reference":[{"key":"bib1","first-page":"1","volume-title":"Proceedings of the AAMAS","author":"Amgoud Leila","year":"2007"},{"key":"bib2","author":"Bajgar Ondrej","year":"2016","journal-title":"CoRR"},{"key":"bib3","doi-asserted-by":"crossref","first-page":"31","DOI":"10.3115\/1219044.1219075","volume-title":"Proceedings of the ACL on Interactive poster and demonstration sessions","author":"Bird Steven","year":"2004"},{"key":"bib4","first-page":"789","volume-title":"Proceedings of the ACL","author":"Chambers Nathanael","year":"2008"},{"key":"bib5","first-page":"2358","volume-title":"Proceedings of the ACL","author":"Chen Danqi","year":"2016"},{"key":"bib6","first-page":"90","volume-title":"Proceedings of the SIGDial","author":"Chen Yu-Hsin","year":"2016"},{"key":"bib7","volume-title":"Proceedings of the AAAI","author":"Chen Zhipeng","year":"2019"},{"key":"bib8","doi-asserted-by":"crossref","unstructured":"Eunsol Choi, He He, Mohit Iyyer, Mark Yatskar, Wen-tau Yih, Yejin Choi, Percy Liang, and Luke Zettlemoyer. 2018. QuAC: Question answering in context. pages 2174\u20132184. Brussels, Belgium.","DOI":"10.18653\/v1\/D18-1241"},{"key":"bib9","first-page":"2580","volume-title":"Proceedings of the AAAI","author":"Clark Peter","year":"2016"},{"key":"bib10","first-page":"593","volume-title":"Proceedings of the ACL","author":"Cui Yiming","year":"2017"},{"key":"bib11","first-page":"1832","volume-title":"Proceedings of the ACL","author":"Dhingra Bhuwan","year":"2017"},{"key":"bib12","author":"Dunn Matthew","year":"2017","journal-title":"CoRR"},{"key":"bib13","first-page":"266","volume-title":"Proceedings of the ACL","author":"Forbes Maxwell","year":"2017"},{"key":"bib14","first-page":"1693","volume-title":"Proceedings of the NIPS","author":"Hermann Karl Moritz","year":"2015"},{"key":"bib15","volume-title":"Proceedings of the ICLR","author":"Hill Felix","year":"2016"},{"key":"bib16","first-page":"213","volume-title":"Proceedings of the COLING","author":"Huang Danqing","year":"2018"},{"key":"bib17","author":"Joshi Mandar","year":"2017","journal-title":"CoRR"},{"key":"bib18","first-page":"252","volume-title":"Proceedings of the NAACL-HLT","author":"Khashabi Daniel","year":"2018"},{"key":"bib19","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00023"},{"key":"bib20","first-page":"785","volume-title":"Proceedings of the EMNLP","author":"Lai Guokun","year":"2017"},{"key":"bib21","author":"Lee Chia-Hsuan","year":"2018","journal-title":"CoRR"},{"key":"bib22","author":"Li Chia-Hsuan","year":"2018","journal-title":"CoRR"},{"key":"bib23","volume-title":"Proceedings of the ICLR","author":"Liu Peter J","year":"2018"},{"key":"bib24","first-page":"329","volume-title":"Proceedings of the ACL","author":"LoBue Peter","year":"2011"},{"key":"bib25","first-page":"2039","volume-title":"Proceedings of the NAACL-HLT","author":"Ma Kaixin","year":"2018"},{"key":"bib26","volume-title":"Proceedings of the EMNLP","author":"Mihaylov Todor","year":"2018"},{"key":"bib27","first-page":"839","volume-title":"Proceedings of the NAACL-HLT","author":"Mostafazadeh Nasrin","year":"2016"},{"key":"bib28","doi-asserted-by":"crossref","first-page":"59","DOI":"10.3138\/cmlr.63.1.59","volume":"63","author":"Nation I","year":"2006","journal-title":"Canadian Modern Language Review"},{"key":"bib29","author":"Nguyen Tri","year":"2016","journal-title":"CoRR"},{"key":"bib30","first-page":"2230","volume-title":"Proceedings of the EMNLP","author":"Onishi Takeshi","year":"2016"},{"key":"bib31","first-page":"747","volume-title":"Proceedings of the SemEval","author":"Ostermann Simon","year":"2018"},{"key":"bib32","first-page":"2825","volume":"12","author":"Pedregosa Fabian","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"bib33","first-page":"1194","volume-title":"Proceedings of the CLEF","author":"Penas Anselmo","year":"2014"},{"key":"bib34","first-page":"2227","volume-title":"Proceedings of the NAACL-HLT","author":"Peters Matthew","year":"2018"},{"key":"bib35","unstructured":"Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. Improving language understanding by generative pre-training. Preprint, available at https:\/\/openai.com\/blog\/language- unsupervised\/."},{"key":"bib36","first-page":"pages 2383\u2013page","volume-title":"Proceedings of the EMNLP","author":"Rajpurkar Pranav","year":"2016"},{"key":"bib37","author":"Reddy Siva","year":"2018","journal-title":"CoRR"},{"key":"bib38","first-page":"193","volume-title":"Proceedings of the EMNLP","author":"Richardson Matthew","year":"2013"},{"key":"bib39","volume-title":"NTCIR","author":"Shibuki Hideyuki","year":"2014"},{"issue":"6","key":"bib40","first-page":"39","volume":"18","author":"Sikes Richard","year":"2007","journal-title":"Multilingual"},{"key":"bib41","first-page":"pages 4444\u2013page","volume-title":"Proceedings of the AAAI","author":"Speer Robyn","year":"2017"},{"key":"bib42","first-page":"1056","volume-title":"Proceedings of the WWW","author":"Tan Chenhao","year":"2015"},{"key":"bib43","author":"Yi Tay","year":"2018","journal-title":"CoRR"},{"key":"bib44","first-page":"191","volume-title":"Proceedings of the RepL4NLP","author":"Trischler Adam","year":"2017"},{"key":"bib45","volume-title":"Proceedings of the Interspeech","author":"Tseng Bo-Hsiang","year":"2016"},{"key":"bib46","first-page":"1064","volume-title":"Proceedings of the EMNLP","author":"Wang Lei","year":"2018"},{"key":"bib47","first-page":"1","volume-title":"Proceedings of the ACL","author":"Wang Shuohang","year":"2018"},{"key":"bib48","first-page":"1744","volume-title":"Proceedings of the ACL","author":"Yih Wen-tau","year":"2013"},{"key":"bib49","volume-title":"Proceedings of the ICLR","author":"Adams Wei Yu","year":"2018"},{"key":"bib50","first-page":"19","volume-title":"Proceedings of the IEEE ICCV","author":"Zhu Yukun","year":"2015"}],"container-title":["Transactions of the Association for Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mitpressjournals.org\/doi\/pdf\/10.1162\/tacl_a_00264","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,12]],"date-time":"2021-03-12T21:39:21Z","timestamp":1615585161000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/43501"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,11]]},"references-count":50,"alternative-id":["10.1162\/tacl_a_00264"],"URL":"https:\/\/doi.org\/10.1162\/tacl_a_00264","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,11]]}}}