{"id":"https://openalex.org/W7166866790","doi":"https://doi.org/10.18653/v1/2026.acl-long.587","title":"Making Large Language Models Efficient Dense Retrievers","display_name":"Making Large Language Models Efficient Dense Retrievers","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166866790","doi":"https://doi.org/10.18653/v1/2026.acl-long.587"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.587","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.587","pdf_url":"https://aclanthology.org/2026.acl-long.587.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.587.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5080310094","display_name":"Yibin Lei","orcid":"https://orcid.org/0009-0007-9558-5548"},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Yibin Lei","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139768000","display_name":"Shwai He","orcid":null},"institutions":[{"id":"https://openalex.org/I66946132","display_name":"University of Maryland, College Park","ror":"https://ror.org/047s2c258","country_code":"US","type":"education","lineage":["https://openalex.org/I66946132"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shwai He","raw_affiliation_strings":["University of Maryland , College Park"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Maryland , College Park","institution_ids":["https://openalex.org/I66946132"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139826301","display_name":"Ang Li","orcid":null},"institutions":[{"id":"https://openalex.org/I66946132","display_name":"University of Maryland, College Park","ror":"https://ror.org/047s2c258","country_code":"US","type":"education","lineage":["https://openalex.org/I66946132"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ang Li","raw_affiliation_strings":["University of Maryland , College Park"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Maryland , College Park","institution_ids":["https://openalex.org/I66946132"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139807547","display_name":"Andrew Yates","orcid":null},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]},{"id":"https://openalex.org/I2802946424","display_name":"Johns Hopkins University Applied Physics Laboratory","ror":"https://ror.org/029pp9z10","country_code":"US","type":"facility","lineage":["https://openalex.org/I145311948","https://openalex.org/I2802946424"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Andrew Yates","raw_affiliation_strings":["Johns Hopkins University , HLTCOE"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Johns Hopkins University , HLTCOE","institution_ids":["https://openalex.org/I145311948","https://openalex.org/I2802946424"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.82529811,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"12898","last_page":"12913"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.4323999881744385,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.4323999881744385,"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/T10028","display_name":"Topic Modeling","score":0.10109999775886536,"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/T11550","display_name":"Text and Document Classification Technologies","score":0.05429999902844429,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.32690000534057617},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.289000004529953},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.2766000032424927},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.2637999951839447},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.25189998745918274}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6309999823570251},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3977999985218048},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.32690000534057617},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.31150001287460327},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2637999951839447},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.25189998745918274},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.23919999599456787},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.23720000684261322}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.587","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.587","pdf_url":"https://aclanthology.org/2026.acl-long.587.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.587","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.587","pdf_url":"https://aclanthology.org/2026.acl-long.587.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2595414520","display_name":"Robust Search with Open-Source LLMs","funder_award_id":"2024.050","funder_id":"https://openalex.org/F4320321800","funder_display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek"}],"funders":[{"id":"https://openalex.org/F4320321800","display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek","ror":"https://ror.org/04jsz6e67"},{"id":"https://openalex.org/F4320322725","display_name":"China Scholarship Council","ror":"https://ror.org/04atp4p48"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166866790.pdf","grobid_xml":"https://content.openalex.org/works/W7166866790.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"work":[1],"has":[2],"shown":[3],"that":[4,113],"directly":[5],"fine-tuning":[6],"large":[7],"language":[8],"models":[9,47,165],"(LLMs)":[10],"for":[11,35,50,98,109],"dense":[12,78],"retrieval":[13,51],"yields":[14],"strong":[15],"performance,":[16],"but":[17],"their":[18],"substantial":[19,142],"parameter":[20],"counts":[21],"make":[22],"them":[23],"computationally":[24],"inefficient.While":[25],"prior":[26],"studies":[27],"have":[28],"revealed":[29],"significant":[30],"layer":[31,74],"redundancy":[32,43,75],"in":[33,76,82,144],"LLMs":[34],"generative":[36,85],"tasks,":[37,52],"it":[38],"remains":[39],"unclear":[40],"whether":[41],"similar":[42],"exists":[44],"when":[45],"these":[46],"are":[48,89,166],"adapted":[49],"which":[53],"require":[54],"encoding":[55],"entire":[56],"sequences":[57],"into":[58],"fixed":[59],"representations":[60],"rather":[61],"than":[62],"generating":[63],"tokens":[64],"iteratively.To":[65],"this":[66,102],"end,":[67],"we":[68,104],"conduct":[69],"a":[70,107,119],"comprehensive":[71],"analysis":[72],"of":[73,154],"LLM-based":[77],"retrievers.We":[79],"find":[80],"that,":[81],"contrast":[83],"to":[84],"settings,":[86],"MLP":[87,116],"layers":[88,95],"substantially":[90],"more":[91],"prunable,":[92],"while":[93,150],"attention":[94],"remain":[96],"critical":[97],"semantic":[99],"aggregation.Building":[100],"on":[101],"insight,":[103],"propose":[105],"EffiR,":[106],"framework":[108],"developing":[110],"efficient":[111],"retrievers":[112],"performs":[114],"large-scale":[115],"compression":[117],"through":[118],"coarseto-fine":[120],"strategy":[121],"(coarse-grained":[122],"depth":[123],"reduction":[124],"followed":[125],"by":[126],"fine-grained":[127],"width":[128],"reduction),":[129],"combined":[130],"with":[131],"retrieval-specific":[132],"fine-tuning.Across":[133],"diverse":[134],"BEIR":[135],"datasets":[136],"and":[137,147,164],"LLM":[138],"backbones,":[139],"EffiR":[140],"achieves":[141],"reductions":[143],"model":[145],"size":[146],"inference":[148],"cost":[149],"preserving":[151],"the":[152],"performance":[153],"full-size":[155],"models.":[156],"1":[157,161],"*":[158],"Equal":[159],"contribution":[160],"Our":[162],"code":[163],"available":[167],"at":[168],"https://github.":[169],"com/":[170]},"counts_by_year":[],"updated_date":"2026-08-25T07:29:55.448023","created_date":"2026-07-02T00:00:00"}