{"id":"https://openalex.org/W3198431451","doi":"https://doi.org/10.1145/3471158.3472233","title":"BERT-based Dense Retrievers Require Interpolation with BM25 for Effective Passage Retrieval","display_name":"BERT-based Dense Retrievers Require Interpolation with BM25 for Effective Passage Retrieval","publication_year":2021,"publication_date":"2021-07-11","ids":{"openalex":"https://openalex.org/W3198431451","doi":"https://doi.org/10.1145/3471158.3472233","mag":"3198431451"},"language":"en","primary_location":{"id":"doi:10.1145/3471158.3472233","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3471158.3472233","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100328263","display_name":"Shuai Wang","orcid":"https://orcid.org/0000-0002-0726-5250"},"institutions":[{"id":"https://openalex.org/I165143802","display_name":"University of Queensland","ror":"https://ror.org/00rqy9422","country_code":"AU","type":"education","lineage":["https://openalex.org/I165143802"]}],"countries":["AU"],"is_corresponding":true,"raw_author_name":"Shuai Wang","raw_affiliation_strings":["The University of Queensland, Brisbane, Australia"],"affiliations":[{"raw_affiliation_string":"The University of Queensland, Brisbane, Australia","institution_ids":["https://openalex.org/I165143802"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012958162","display_name":"Shengyao Zhuang","orcid":"https://orcid.org/0000-0002-6711-0955"},"institutions":[{"id":"https://openalex.org/I165143802","display_name":"University of Queensland","ror":"https://ror.org/00rqy9422","country_code":"AU","type":"education","lineage":["https://openalex.org/I165143802"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Shengyao Zhuang","raw_affiliation_strings":["The University of Queensland, Brisbane, Australia"],"affiliations":[{"raw_affiliation_string":"The University of Queensland, Brisbane, Australia","institution_ids":["https://openalex.org/I165143802"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5076031002","display_name":"Guido Zuccon","orcid":"https://orcid.org/0000-0003-0271-5563"},"institutions":[{"id":"https://openalex.org/I165143802","display_name":"University of Queensland","ror":"https://ror.org/00rqy9422","country_code":"AU","type":"education","lineage":["https://openalex.org/I165143802"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Guido Zuccon","raw_affiliation_strings":["The University of Queensland, Brisbane, Australia"],"affiliations":[{"raw_affiliation_string":"The University of Queensland, Brisbane, Australia","institution_ids":["https://openalex.org/I165143802"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5100328263"],"corresponding_institution_ids":["https://openalex.org/I165143802"],"apc_list":null,"apc_paid":null,"fwci":6.1581,"has_fulltext":false,"cited_by_count":60,"citation_normalized_percentile":{"value":0.96951343,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"317","last_page":"324"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9994000196456909,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10286","display_name":"Information Retrieval and Search Behavior","score":0.9973000288009644,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7909308075904846},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.6174470782279968},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6044155359268188},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.5668094754219055},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5637551546096802},{"id":"https://openalex.org/keywords/relevance","display_name":"Relevance (law)","score":0.5261691808700562},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.515445351600647},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5154098272323608},{"id":"https://openalex.org/keywords/interpolation","display_name":"Interpolation (computer graphics)","score":0.45352017879486084}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7909308075904846},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6174470782279968},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6044155359268188},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.5668094754219055},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5637551546096802},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.5261691808700562},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.515445351600647},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5154098272323608},{"id":"https://openalex.org/C137800194","wikidata":"https://www.wikidata.org/wiki/Q11713455","display_name":"Interpolation (computer graphics)","level":3,"score":0.45352017879486084},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3471158.3472233","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3471158.3472233","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4399999976158142,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W2004545875","https://openalex.org/W2117473841","https://openalex.org/W2155482025","https://openalex.org/W2899154813","https://openalex.org/W2945127593","https://openalex.org/W2963341956","https://openalex.org/W2998702515","https://openalex.org/W3021397474","https://openalex.org/W3021779606","https://openalex.org/W3038572442","https://openalex.org/W3082274269","https://openalex.org/W3099384026","https://openalex.org/W3099446234","https://openalex.org/W3099700870","https://openalex.org/W3104657626","https://openalex.org/W3130740619","https://openalex.org/W3145630588","https://openalex.org/W3147292006","https://openalex.org/W3148323213","https://openalex.org/W3155114168","https://openalex.org/W3155895380","https://openalex.org/W3175111331","https://openalex.org/W4252076394","https://openalex.org/W4323063355"],"related_works":["https://openalex.org/W2468279273","https://openalex.org/W2354198838","https://openalex.org/W1989130879","https://openalex.org/W2103419012","https://openalex.org/W2988126442","https://openalex.org/W1974414866","https://openalex.org/W2150136235","https://openalex.org/W2053591227","https://openalex.org/W2041353081","https://openalex.org/W2581240705"],"abstract_inverted_index":{"The":[0],"integration":[1],"of":[2,138],"pre-trained":[3],"deep":[4,38],"language":[5,39],"models,":[6],"such":[7,37,82],"as":[8,67,83,191,197],"BERT,":[9],"into":[10,168],"retrieval":[11,29],"and":[12,77,95,141,160,166,183,221],"ranking":[13],"pipelines":[14],"has":[15,56],"shown":[16,58],"to":[17,48,71,91,98,121,218,265],"provide":[18],"large":[19],"effectiveness":[20],"gains":[21,223],"over":[22],"traditional":[23],"bag-of-words":[24,80,96],"models":[25,40],"in":[26,61,189,250],"the":[27,32,100,105,114,116,124,136,149,222,226,259],"passage":[28],"task.":[30],"However,":[31],"best":[33],"setup":[34],"for":[35,206,214],"integrating":[36],"is":[41,46,85,88,212,235,263],"still":[42],"unclear.":[43],"When":[44],"BERT":[45,52,63,93,106,117,150,207],"used":[47,112,188],"re-rank":[49],"passages":[50],"(i.e.,":[51],"re-ranker),":[53],"previous":[54,145],"work":[55,146],"empirically":[57],"that,":[59,204],"while":[60],"practice":[62,190],"re-ranker":[64,94,107,118],"cannot":[65],"act":[66],"initial":[68],"retriever":[69],"due":[70],"BERT's":[72,176],"high":[73,177],"query":[74,181],"time":[75],"costs,":[76],"thus":[78,185],"a":[79,256],"model":[81],"BM25":[84,140,211,262],"required.":[86],"It":[87],"not":[89],"necessary":[90,213],"interpolate":[92],"scores":[97,108],"generate":[99],"final":[101],"ranking.":[102],"In":[103,130],"fact,":[104],"alone":[109],"can":[110,184],"be":[111,186],"by":[113,128,225],"re-ranker:":[115],"score":[119],"appears":[120],"already":[122],"capture":[123],"relevance":[125,245,253],"signal":[126],"provided":[127,224],"BM25.":[129],"this":[131,234],"paper,":[132],"we":[133,154],"further":[134],"investigate":[135],"topic":[137],"interpolating":[139],"BERT-based":[142,156,171,215],"rankers.":[143],"Unlike":[144],"that":[147,258],"considered":[148],"re-ranker,":[151,208],"however,":[152],"here":[153],"consider":[155],"dense":[157,216,237],"retrievers":[158,163,217,238],"(RepBERT":[159],"ANCE).":[161],"Dense":[162],"encode":[164],"queries":[165],"documents":[167],"low":[169],"dimensional":[170],"embeddings.":[172],"These":[173],"methods":[174],"overcome":[175],"computational":[178],"costs":[179],"at":[180,242],"time,":[182],"feasibly":[187],"whole-collection":[192],"retrievers,":[193],"rather":[194],"than":[195],"just":[196],"re-rankers.":[198],"Our":[199],"novel":[200],"empirical":[201],"findings":[202],"suggest":[203],"unlike":[205],"interpolation":[209,227,260],"with":[210,261],"perform":[219],"effectively;":[220],"are":[228,239],"significant.":[229],"Further":[230],"analysis":[231],"reveals":[232],"why":[233],"so:":[236],"very":[240],"effective":[241],"encoding":[243],"strong":[244],"signals,":[246],"but":[247],"they":[248],"fail":[249],"identifying":[251],"weaker":[252],"signals":[254],"--":[255],"task":[257],"able":[264],"make":[266],"up":[267],"for.":[268]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":15},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":17},{"year":2022,"cited_by_count":14},{"year":2021,"cited_by_count":2}],"updated_date":"2026-04-05T17:49:38.594831","created_date":"2025-10-10T00:00:00"}