{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T14:32:11Z","timestamp":1775485931879,"version":"3.50.1"},"reference-count":53,"publisher":"MIT Press","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Transactions of the Association for Computational Linguistics"],"published-print":{"date-parts":[[2020,12]]},"abstract":"<jats:p> Recent work has presented intriguing results examining the knowledge contained in language models (LMs) by having the LM fill in the blanks of prompts such as \u201c Obama is a __ by profession\u201d. These prompts are usually manually created, and quite possibly sub-optimal; another prompt such as \u201c Obama worked as a __ \u201d may result in more accurately predicting the correct profession. Because of this, given an inappropriate prompt, we might fail to retrieve facts that the LM does know, and thus any given prompt only provides a lower bound estimate of the knowledge contained in an LM. In this paper, we attempt to more accurately estimate the knowledge contained in LMs by automatically discovering better prompts to use in this querying process. Specifically, we propose mining-based and paraphrasing-based methods to automatically generate high-quality and diverse prompts, as well as ensemble methods to combine answers from different prompts. Extensive experiments on the LAMA benchmark for extracting relational knowledge from LMs demonstrate that our methods can improve accuracy from 31.1% to 39.6%, providing a tighter lower bound on what LMs know. We have released the code and the resulting LM Prompt And Query Archive (LPAQA) at https:\/\/github.com\/jzbjyb\/LPAQA . <\/jats:p>","DOI":"10.1162\/tacl_a_00324","type":"journal-article","created":{"date-parts":[[2020,7,20]],"date-time":"2020-07-20T18:01:16Z","timestamp":1595268076000},"page":"423-438","source":"Crossref","is-referenced-by-count":592,"title":["How Can We Know What Language Models Know?"],"prefix":"10.1162","volume":"8","author":[{"given":"Zhengbao","family":"Jiang","sequence":"first","affiliation":[{"name":"Language Technologies Institute, Carnegie Mellon University."}]},{"given":"Frank F.","family":"Xu","sequence":"additional","affiliation":[{"name":"Language Technologies Institute, Carnegie Mellon University."}]},{"given":"Jun","family":"Araki","sequence":"additional","affiliation":[{"name":"Bosch Research North America."}]},{"given":"Graham","family":"Neubig","sequence":"additional","affiliation":[{"name":"Language Technologies Institute, Carnegie Mellon University."}]}],"member":"281","reference":[{"key":"bib1","first-page":"85","volume-title":"Proceedings of the Fifth ACM Conference on Digital Libraries, June 2-7, 2000, San Antonio, TX, USA","author":"Agichtein Eugene","year":"2000"},{"key":"bib2","author":"Ahn Sungjin","year":"2016","journal-title":"CoRR"},{"key":"bib3","doi-asserted-by":"crossref","first-page":"2895","DOI":"10.18653\/v1\/P19-1279","volume-title":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","author":"Soares Livio Baldini","year":"2019"},{"key":"bib4","first-page":"2670","volume-title":"IJCAI 2007, Proceedings of the 20th International Joint Conference on Artificial Intelligence, Hyderabad, India, January 6-12, 2007","author":"Banko Michele","year":"2007"},{"key":"bib5","doi-asserted-by":"crossref","first-page":"861","DOI":"10.18653\/v1\/P17-1080","volume-title":"Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","author":"Belinkov Yonatan","year":"2017"},{"key":"bib6","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00254"},{"key":"bib7","first-page":"674","volume-title":"Proceedings of ACL-08: HLT","author":"Bhagat Rahul","year":"2008"},{"key":"bib8","volume-title":"Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI)","author":"Bouraoui Zied","year":"2020"},{"issue":"1","key":"bib9","first-page":"1:1","volume":"44","author":"Carpineto Claudio","year":"2012","journal-title":"ACM, Computing Surveys"},{"key":"bib10","first-page":"3079","volume-title":"Advances in Neural Information Processing Systems 28: Annual Conference on Neural Information Processing Systems 2015, December 7-12, 2015, Montreal, Quebec, Canada","author":"Dai Andrew M.","year":"2015"},{"key":"bib11","first-page":"4171","volume-title":"Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, 2019, Volume 1 (Long and Short Papers)","author":"Devlin Jacob","year":"2019"},{"key":"bib12","doi-asserted-by":"crossref","first-page":"31","DOI":"10.18653\/v1\/P18-2006","volume-title":"Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)","author":"Ebrahimi Javid","year":"2018"},{"key":"bib13","volume-title":"Proceedings of the Eleventh International Conference on Language Resources and Evaluation, LREC 2018, Miyazaki, Japan, May 7-12, 2018","author":"ElSahar Hady","year":"2018"},{"key":"bib14","first-page":"1535","volume-title":"Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, EMNLP 2011, 27-31 July 2011, John McIntyre Conference Centre, Edinburgh, UK, A meeting of SIGDAT, a Special Interest Group of the ACL","author":"Fader Anthony","year":"2011"},{"key":"bib15","first-page":"103","volume-title":"Proceedings of EAMT","author":"Gamon Michael","year":"2005"},{"key":"bib16","first-page":"6114","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Ghazvininejad Marjan","year":"2019"},{"key":"bib17","author":"Goldberg Yoav","year":"2019","journal-title":"CoRR"},{"key":"bib18","volume-title":"Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI)","author":"Hayashi Hiroaki","year":"2020"},{"key":"bib19","first-page":"4129","volume-title":"Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, 2019, Volume 1 (Long and Short Papers)","author":"Hewitt John","year":"2019"},{"key":"bib20","first-page":"146","volume-title":"Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing","author":"Hoang Cong Duy Vu","year":"2017"},{"key":"bib21","first-page":"5962","volume-title":"Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1: Long Papers","author":"Logan Robert L.","year":"2019"},{"key":"bib22","first-page":"3651","volume-title":"Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1: Long Papers","author":"Jawahar Ganesh","year":"2019"},{"key":"bib23","volume-title":"3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings","author":"Kingma Diederik P.","year":"2015"},{"key":"bib24","first-page":"110","volume-title":"NAACL HLT 2016, The 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, San Diego California, USA, June 12-17, 2016","author":"Li Jiwei","year":"2016"},{"key":"bib25","author":"Li Jiwei","year":"2016","journal-title":"CoRR"},{"key":"bib26","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00115"},{"key":"bib27","first-page":"881","volume-title":"Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers","author":"Mallinson Jonathan","year":"2017"},{"key":"bib28","author":"McCann Bryan","year":"2018","journal-title":"CoRR"},{"key":"bib29","first-page":"51","volume-title":"Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning, CoNLL 2016, Berlin, Germany, August 11-12, 2016","author":"Melamud Oren","year":"2016"},{"key":"bib30","volume-title":"6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 - May 3, 2018, Conference Track Proceedings","author":"Melis G\u00e1bor","year":"2018"},{"key":"bib31","volume-title":"6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 - May 3, 2018, Conference Track Proceedings","author":"Merity Stephen","year":"2018"},{"key":"bib32","doi-asserted-by":"crossref","first-page":"234","DOI":"10.1109\/SLT.2012.6424228","volume-title":"2012 IEEE Spoken Language Technology Workshop (SLT)","author":"Mikolov Tomas","year":"2012"},{"key":"bib33","first-page":"314","volume-title":"Proceedings of the Fourth Conference on Machine Translation, WMT 2019, Florence, Italy, August 1-2, 2019 - Volume 2: Shared Task Papers, Day 1","author":"Ng Nathan","year":"2019"},{"key":"bib34","first-page":"2227","volume-title":"Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2018, New Orleans, Louisiana, USA, June 1-6, 2018, Volume 1 (Long Papers)","author":"Peters Matthew E.","year":"2018"},{"key":"bib35","doi-asserted-by":"crossref","first-page":"43","DOI":"10.18653\/v1\/D19-1005","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Peters Matthew E.","year":"2019"},{"key":"bib36","doi-asserted-by":"crossref","first-page":"2463","DOI":"10.18653\/v1\/D19-1250","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Petroni Fabio","year":"2019"},{"key":"bib37","author":"P\u00f6rner Nina","year":"2019","journal-title":"CoRR"},{"issue":"8","key":"bib38","volume":"1","author":"Radford Alec","year":"2019","journal-title":"OpenAI Blog"},{"key":"bib39","doi-asserted-by":"crossref","first-page":"4932","DOI":"10.18653\/v1\/P19-1487","volume-title":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","author":"Rajani Nazneen Fatema","year":"2019"},{"key":"bib40","first-page":"41","volume-title":"Proceedings of the 40th annual meeting on association for computational linguistics","author":"Ravichandran Deepak","year":"2002"},{"key":"bib41","volume-title":"11th Conference of the European Chapter of the Association for Computational Linguistics","author":"Romano Lorenza","year":"2006"},{"key":"bib42","first-page":"3027","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","volume":"33","author":"Sap Maarten","year":"2019"},{"key":"bib43","volume-title":"Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016, August 7-12, 2016, Berlin, Germany, Volume 1: Long Papers","author":"Sennrich Rico","year":"2016"},{"key":"bib44","first-page":"1526","volume-title":"Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing","author":"Shi Xing","year":"2016"},{"key":"bib45","first-page":"1249","volume-title":"Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2017, Valencia, Spain, April 3-7, 2017, Volume 1: Long Papers","author":"Smith Noah A.","year":"2017"},{"key":"bib46","first-page":"4593","volume-title":"Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1: Long Papers","author":"Tenney Ian","year":"2019"},{"key":"bib47","volume-title":"7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019","author":"Tenney Ian","year":"2019"},{"key":"bib48","first-page":"1499","volume-title":"Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015, Lisbon, Portugal, September 17-21, 2015","author":"Toutanova Kristina","year":"2015"},{"key":"bib49","author":"Trinh Trieu H.","year":"2018","journal-title":"CoRR"},{"key":"bib50","doi-asserted-by":"crossref","first-page":"2153","DOI":"10.18653\/v1\/D19-1221","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Wallace Eric","year":"2019"},{"key":"bib51","first-page":"1850","volume-title":"Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017, Copenhagen, Denmark, September 9-11, 2017","author":"Yang Zichao","year":"2017"},{"key":"bib52","first-page":"1441","volume-title":"Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1: Long Papers","author":"Zhang Zhengyan","year":"2019"},{"key":"bib53","unstructured":"Geoffrey Zweig and Christopher J. C. Burges. 2011. The Microsoft Research sentence completion challenge. Microsoft Research, Redmond, WA, USA, Technical Report MSR-TR-2011-129."}],"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_00324","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,12]],"date-time":"2021-03-12T21:39:41Z","timestamp":1615585181000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/96460"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,12]]},"references-count":53,"alternative-id":["10.1162\/tacl_a_00324"],"URL":"https:\/\/doi.org\/10.1162\/tacl_a_00324","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,12]]}}}