{"id":"https://openalex.org/W7160619281","doi":"https://doi.org/10.48550/arxiv.2605.05532","title":"A Few Good Clauses: Comparing LLMs vs Domain-Trained Small Language Models on Structured Contract Extraction","display_name":"A Few Good Clauses: Comparing LLMs vs Domain-Trained Small Language Models on Structured Contract Extraction","publication_year":2026,"publication_date":"2026-05-07","ids":{"openalex":"https://openalex.org/W7160619281","doi":"https://doi.org/10.48550/arxiv.2605.05532"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.05532","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05532","pdf_url":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.05532","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020476039","display_name":"Nicole Lincoln","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lincoln, Nicole","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093826663","display_name":"Nick Whitehouse","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Whitehouse, Nick","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008026205","display_name":"Jaron Mar","orcid":"https://orcid.org/0000-0002-2393-6229"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mar, Jaron","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5113166433","display_name":"Rivindu Perera","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Perera, Rivindu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13643","display_name":"Artificial Intelligence in Law","score":0.6815000176429749,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T13643","display_name":"Artificial Intelligence in Law","score":0.6815000176429749,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.03830000013113022,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T13851","display_name":"Law, AI, and Intellectual Property","score":0.022299999371170998,"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/frontier","display_name":"Frontier","score":0.6625999808311462},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.59170001745224},{"id":"https://openalex.org/keywords/hallucinating","display_name":"Hallucinating","score":0.5759999752044678},{"id":"https://openalex.org/keywords/aggregate","display_name":"Aggregate (composite)","score":0.5256999731063843},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.4772999882698059},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.46299999952316284},{"id":"https://openalex.org/keywords/macro","display_name":"Macro","score":0.4498000144958496}],"concepts":[{"id":"https://openalex.org/C2778571376","wikidata":"https://www.wikidata.org/wiki/Q1355821","display_name":"Frontier","level":2,"score":0.6625999808311462},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.59170001745224},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5864999890327454},{"id":"https://openalex.org/C2911011789","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Hallucinating","level":2,"score":0.5759999752044678},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.5256999731063843},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.4772999882698059},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.46299999952316284},{"id":"https://openalex.org/C166955791","wikidata":"https://www.wikidata.org/wiki/Q629579","display_name":"Macro","level":2,"score":0.4498000144958496},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4471000134944916},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44130000472068787},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4081999957561493},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.37770000100135803},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.36309999227523804},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3206000030040741},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.30649998784065247},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.3041999936103821},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.29019999504089355},{"id":"https://openalex.org/C4725764","wikidata":"https://www.wikidata.org/wiki/Q844704","display_name":"Extraction (chemistry)","level":2,"score":0.2833999991416931},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.28110000491142273},{"id":"https://openalex.org/C108759981","wikidata":"https://www.wikidata.org/wiki/Q788590","display_name":"Authorization","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C162118730","wikidata":"https://www.wikidata.org/wiki/Q1128453","display_name":"Actuarial science","level":1,"score":0.2721000015735626},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2705000042915344},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.26100000739097595},{"id":"https://openalex.org/C42475967","wikidata":"https://www.wikidata.org/wiki/Q194292","display_name":"Operations research","level":1,"score":0.2590000033378601},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25279998779296875},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.05532","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05532","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.05532","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05532","pdf_url":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9","score":0.6731138229370117}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"evaluates":[2],"whether":[3],"a":[4,29,53,59],"domain":[5,33],"trained":[6],"Small":[7],"Language":[8,15],"Model":[9],"(SLM)":[10],"can":[11],"outperform":[12],"frontier":[13,40,75],"Large":[14],"Models":[16],"on":[17],"structured":[18],"contract":[19],"extraction":[20],"at":[21],"radically":[22],"lower":[23],"cost.":[24],"We":[25],"test":[26],"Olava":[27,42],"Extract,":[28],"self":[30],"hosted":[31,122,148],"legal":[32,95,114],"Mixture":[34],"of":[35,56,62],"Experts":[36],"model,":[37],"against":[38],"five":[39],"models.":[41,123],"Extract":[43],"achieved":[44,80],"the":[45,50,74,81,119,128],"strongest":[46],"aggregate":[47],"performance":[48],"in":[49,94],"study,":[51],"with":[52,73],"macro":[54],"F1":[55,61],"0.812":[57],"and":[58,88,102,146],"micro":[60],"0.842,":[63],"while":[64],"reducing":[65],"inference":[66],"cost":[67],"by":[68],"78%":[69],"to":[70,139],"97%":[71],"compared":[72],"models":[76],"tested.":[77],"It":[78],"also":[79],"highest":[82],"precision":[83],"scores,":[84],"producing":[85],"fewer":[86],"hallucinated":[87],"unsupported":[89],"extractions,":[90],"an":[91],"important":[92],"distinction":[93],"workflows":[96],"where":[97],"hallucinations":[98],"create":[99],"operational":[100],"risk":[101],"downstream":[103],"review":[104],"burden.":[105],"The":[106],"findings":[107],"shows":[108],"that":[109,130],"high":[110],"performing,":[111],"human":[112],"comparable":[113],"AI":[115,134],"no":[116],"longer":[117],"requires":[118],"largest":[120],"externally":[121],"More":[124],"broadly,":[125],"they":[126],"challenge":[127],"assumption":[129],"commercially":[131],"valuable":[132],"enterprise":[133],"capability":[135],"must":[136],"remain":[137],"tied":[138],"ever":[140],"larger":[141],"models,":[142],"massive":[143],"infrastructure":[144],"expenditure,":[145],"centrally":[147],"providers.":[149]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-09T00:00:00"}