{"id":"https://openalex.org/W4411403485","doi":"https://doi.org/10.1145/3725337","title":"<scp>Pneuma</scp> : Leveraging LLMs for Tabular Data Representation and Retrieval in an End-to-End System","display_name":"<scp>Pneuma</scp> : Leveraging LLMs for Tabular Data Representation and Retrieval in an End-to-End System","publication_year":2025,"publication_date":"2025-06-17","ids":{"openalex":"https://openalex.org/W4411403485","doi":"https://doi.org/10.1145/3725337"},"language":"en","primary_location":{"id":"doi:10.1145/3725337","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725337","pdf_url":null,"source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Management of Data","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1145/3725337","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5118331004","display_name":"Muhammad Imam Luthfi Balaka","orcid":null},"institutions":[{"id":"https://openalex.org/I29617571","display_name":"University of Indonesia","ror":"https://ror.org/0116zj450","country_code":"ID","type":"education","lineage":["https://openalex.org/I29617571"]}],"countries":["ID"],"is_corresponding":false,"raw_author_name":"Muhammad Imam Luthfi Balaka","raw_affiliation_strings":["University of Indonesia, Depok, Indonesia"],"raw_orcid":"https://orcid.org/0009-0001-5324-7758","affiliations":[{"raw_affiliation_string":"University of Indonesia, Depok, Indonesia","institution_ids":["https://openalex.org/I29617571"]}]},{"author_position":"middle","author":{"id":null,"display_name":"David Alexander","orcid":"https://orcid.org/0009-0008-6239-9092"},"institutions":[{"id":"https://openalex.org/I29617571","display_name":"University of Indonesia","ror":"https://ror.org/0116zj450","country_code":"ID","type":"education","lineage":["https://openalex.org/I29617571"]}],"countries":["ID"],"is_corresponding":false,"raw_author_name":"David Alexander","raw_affiliation_strings":["University of Indonesia, Depok, Indonesia"],"raw_orcid":"https://orcid.org/0009-0008-6239-9092","affiliations":[{"raw_affiliation_string":"University of Indonesia, Depok, Indonesia","institution_ids":["https://openalex.org/I29617571"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038065917","display_name":"Qiming Wang","orcid":"https://orcid.org/0009-0004-6554-4161"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qiming Wang","raw_affiliation_strings":["The University of Chicago, Chicago, USA"],"raw_orcid":"https://orcid.org/0009-0004-6554-4161","affiliations":[{"raw_affiliation_string":"The University of Chicago, Chicago, USA","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063857016","display_name":"Yue Gong","orcid":"https://orcid.org/0009-0002-8646-473X"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yue Gong","raw_affiliation_strings":["The University of Chicago, Chicago, USA"],"raw_orcid":"https://orcid.org/0009-0002-8646-473X","affiliations":[{"raw_affiliation_string":"The University of Chicago, Chicago, USA","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008983457","display_name":"Adila Alfa Krisnadhi","orcid":"https://orcid.org/0000-0003-0745-6804"},"institutions":[{"id":"https://openalex.org/I29617571","display_name":"University of Indonesia","ror":"https://ror.org/0116zj450","country_code":"ID","type":"education","lineage":["https://openalex.org/I29617571"]}],"countries":["ID"],"is_corresponding":false,"raw_author_name":"Adila Krisnadhi","raw_affiliation_strings":["University of Indonesia, Depok, Indonesia"],"raw_orcid":"https://orcid.org/0000-0003-0745-6804","affiliations":[{"raw_affiliation_string":"University of Indonesia, Depok, Indonesia","institution_ids":["https://openalex.org/I29617571"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003690515","display_name":"Raul Castro Fernandez","orcid":"https://orcid.org/0000-0001-7675-6080"},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Raul Castro Fernandez","raw_affiliation_strings":["The University of Chicago, Chicago, USA"],"raw_orcid":"https://orcid.org/0000-0001-7675-6080","affiliations":[{"raw_affiliation_string":"The University of Chicago, Chicago, USA","institution_ids":["https://openalex.org/I40347166"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":10.3899,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.9792437,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":"3","issue":"3","first_page":"1","last_page":"28"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9995999932289124,"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":0.9995999932289124,"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":0.996399998664856,"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/T11719","display_name":"Data Quality and Management","score":0.9919999837875366,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7133005857467651},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.6177017688751221},{"id":"https://openalex.org/keywords/table","display_name":"Table (database)","score":0.6042748689651489},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.3277128338813782}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7133005857467651},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.6177017688751221},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.6042748689651489},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3277128338813782}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3725337","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725337","pdf_url":null,"source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Management of Data","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2504.09207","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.09207","pdf_url":"https://arxiv.org/pdf/2504.09207","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3725337","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725337","pdf_url":null,"source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Management of Data","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.5899999737739563,"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W2096541451","https://openalex.org/W2788550262","https://openalex.org/W2795089200","https://openalex.org/W2914304175","https://openalex.org/W2926805670","https://openalex.org/W2963469388","https://openalex.org/W2990138404","https://openalex.org/W3004210300","https://openalex.org/W3008881932","https://openalex.org/W3027879771","https://openalex.org/W3035140194","https://openalex.org/W3035231859","https://openalex.org/W3081749234","https://openalex.org/W3099700870","https://openalex.org/W3099839495","https://openalex.org/W3153032435","https://openalex.org/W3157891451","https://openalex.org/W3167748596","https://openalex.org/W3170657538","https://openalex.org/W3196904276","https://openalex.org/W3212368439","https://openalex.org/W4210451781","https://openalex.org/W4252076394","https://openalex.org/W4281845057","https://openalex.org/W4285045050","https://openalex.org/W4312600202","https://openalex.org/W4377138005","https://openalex.org/W4385270264","https://openalex.org/W4385281886","https://openalex.org/W4386576685","https://openalex.org/W4389609672","https://openalex.org/W4399208468","https://openalex.org/W4399397688","https://openalex.org/W4400433884","https://openalex.org/W4400910190","https://openalex.org/W4402670856"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4394360958"],"abstract_inverted_index":{"Finding":[0],"relevant":[1,28,84],"tables":[2],"among":[3],"databases,":[4,208],"lakes,":[5],"and":[6,61,112,129,144,153,175,211,229,235],"repositories":[7],"is":[8,27,46],"the":[9,43,52,179],"first":[10],"step":[11],"in":[12,103,233],"extracting":[13],"value":[14],"from":[15],"data.":[16,85,133,213],"Such":[17],"a":[18,25,30,94,121],"task":[19],"remains":[20],"difficult":[21],"because":[22],"assessing":[23],"whether":[24],"table":[26,142,145,148,162,197,222],"to":[29,51,82,99,127,156,183],"problem":[31],"does":[32],"not":[33],"always":[34],"depend":[35],"only":[36],"on":[37,42,71,200],"its":[38],"content":[39],"but":[40],"also":[41],"context,":[44],"which":[45],"usually":[47],"tribal":[48],"knowledge":[49],"known":[50],"individual":[53],"or":[54,74],"team.":[55],"While":[56],"tools":[57],"like":[58],"data":[59,63,159],"catalogs":[60],"academic":[62],"discovery":[64,108,198],"systems":[65,224],"target":[66],"this":[67,115],"problem,":[68],"they":[69],"rely":[70],"keyword":[72],"search":[73,223,228],"more":[75,109],"complex":[76],"interfaces,":[77],"limiting":[78],"non-technical":[79],"users'":[80],"ability":[81],"find":[83],"The":[86],"advent":[87],"of":[88,181],"large":[89,136],"language":[90,137],"models":[91,138],"(LLMs)":[92,139],"offers":[93],"unique":[95],"opportunity":[96],"for":[97,140],"users":[98],"ask":[100],"questions":[101],"directly":[102],"natural":[104],"language,":[105],"making":[106],"dataset":[107],"intuitive,":[110],"accessible,":[111],"efficient.":[113],"In":[114],"paper,":[116],"we":[117,191],"introduce":[118],"Pneuma":[119,134,150,164,189,218],",":[120,190],"retrieval-augmented":[122],"generation":[123],"(RAG)":[124],"system":[125],"designed":[126],"efficiently":[128],"effectively":[130],"discover":[131],"tabular":[132],"leverages":[135],"both":[141,182],"representation":[143],"retrieval.":[146],"For":[147,161],"representation,":[149],"preserves":[151],"schema":[152],"row-level":[154],"information":[155,169],"ensure":[157],"comprehensive":[158,193],"understanding.":[160],"retrieval,":[163],"augments":[165],"LLMs":[166],"with":[167],"traditional":[168],"retrieval":[170,185],"techniques,":[171],"such":[172],"as":[173,226],"full-text":[174,227],"vector":[176],"search,":[177],"harnessing":[178],"strengths":[180],"improve":[184],"performance.":[186],"To":[187],"evaluate":[188],"generate":[192],"benchmarks":[194],"that":[195,217],"simulate":[196],"workload":[199],"six":[201],"real-world":[202],"datasets":[203],"including":[204],"enterprise":[205],"data,":[206,210],"scientific":[207],"warehousing":[209],"open":[212],"Our":[214],"results":[215],"demonstrate":[216],"outperforms":[219],"widely":[220],"used":[221],"(such":[225],"state-of-the-art":[230],"RAG":[231],"systems)":[232],"accuracy":[234],"resource":[236],"efficiency.":[237]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":3}],"updated_date":"2026-07-16T13:24:37.021932","created_date":"2025-06-19T00:00:00"}