{"id":"https://openalex.org/W7222765853","doi":"https://doi.org/10.64898/2026.10.15.26364476","title":"Multimodal Artificial Intelligence for Cause-of-Death Assignment From Verbal Autopsy Records: A National Benchmark Study in South Africa","display_name":"Multimodal Artificial Intelligence for Cause-of-Death Assignment From Verbal Autopsy Records: A National Benchmark Study in South Africa","publication_year":2026,"publication_date":"2026-10-16","ids":{"openalex":"https://openalex.org/W7222765853","doi":"https://doi.org/10.64898/2026.10.15.26364476"},"language":"en","primary_location":{"id":"doi:10.64898/2026.10.15.26364476","doi":"https://doi.org/10.64898/2026.10.15.26364476","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.10.15.26364476","pdf_url":"https://www.medrxiv.org/content/medrxiv/early/2026/10/16/2026.10.15.26364476.full.pdf","source":{"id":"https://openalex.org/S3005729997","display_name":"medRxiv","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":["https://openalex.org/P4310315909"],"host_organization_lineage_names":["Cold Spring Harbor Laboratory Press"],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"medRxiv","raw_type":"posted-content"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.medrxiv.org/content/medrxiv/early/2026/10/16/2026.10.15.26364476.full.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5036299936","display_name":"Yue Chu","orcid":"https://orcid.org/0000-0001-8671-5362","observed_orcids":["https://orcid.org/0000-0001-8671-5362"]},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yue Chu","raw_affiliation_strings":["Department of Sociology, The Ohio State University","Institute for Population Research, The Ohio State University"],"raw_orcid":"https://orcid.org/0000-0001-8671-5362","affiliations":[{"raw_affiliation_string":"Department of Sociology, The Ohio State University","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"Institute for Population Research, The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034218648","display_name":"Zhenke Wu","orcid":"https://orcid.org/0000-0001-7582-669X","observed_orcids":["https://orcid.org/0000-0001-7582-669X"]},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhenke Wu","raw_affiliation_strings":["Department of Biostatistics, University of Michigan"],"raw_orcid":"https://orcid.org/0000-0001-7582-669X","affiliations":[{"raw_affiliation_string":"Department of Biostatistics, University of Michigan","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015326615","display_name":"Tyler H. McCormick","orcid":"https://orcid.org/0000-0002-6490-1129","observed_orcids":["https://orcid.org/0000-0002-6490-1129"]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tyler H. McCormick","raw_affiliation_strings":["Department of Statistics, University of Washington"],"raw_orcid":"https://orcid.org/0000-0002-6490-1129","affiliations":[{"raw_affiliation_string":"Department of Statistics, University of Washington","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083814134","display_name":"Zehang Richard Li","orcid":"https://orcid.org/0000-0001-9551-9638","observed_orcids":["https://orcid.org/0000-0001-9551-9638"]},"institutions":[{"id":"https://openalex.org/I185103710","display_name":"University of California, Santa Cruz","ror":"https://ror.org/03s65by71","country_code":"US","type":"education","lineage":["https://openalex.org/I185103710"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zehang Richard Li","raw_affiliation_strings":["Department of Statistics, University of California Santa Cruz"],"raw_orcid":"https://orcid.org/0000-0001-9551-9638","affiliations":[{"raw_affiliation_string":"Department of Statistics, University of California Santa Cruz","institution_ids":["https://openalex.org/I185103710"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063372237","display_name":"Samuel J. Clark","orcid":"https://orcid.org/0000-0002-4929-6231","observed_orcids":["https://orcid.org/0000-0002-4929-6231"]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]},{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Samuel J. Clark","raw_affiliation_strings":["Department of Sociology, The Ohio State University","Department of Statistics, University of Washington","Institute for Population Research, The Ohio State University"],"raw_orcid":"https://orcid.org/0000-0002-4929-6231","affiliations":[{"raw_affiliation_string":"Department of Sociology, The Ohio State University","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"Department of Statistics, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"Institute for Population Research, The Ohio State University","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5036299936"],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T11396","display_name":"Artificial Intelligence in Healthcare","score":0.4936999976634979,"subfield":{"id":"https://openalex.org/subfields/3605","display_name":"Health Information Management"},"field":{"id":"https://openalex.org/fields/36","display_name":"Health Professions"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11396","display_name":"Artificial Intelligence in Healthcare","score":0.4936999976634979,"subfield":{"id":"https://openalex.org/subfields/3605","display_name":"Health Information Management"},"field":{"id":"https://openalex.org/fields/36","display_name":"Health Professions"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.3472000062465668,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.015300000086426735,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/cause-of-death-assignment","display_name":"cause of death assignment","score":0.9039999842643738},{"id":"https://openalex.org/keywords/verbal-autopsy","display_name":"verbal autopsy","score":0.9879999756813049},{"id":"https://openalex.org/keywords/multimodal-learning","display_name":"multimodal learning","score":0.8299999833106995},{"id":"https://openalex.org/keywords/decision-level-fusion","display_name":"decision fusion","score":0.5410000085830688},{"id":"https://openalex.org/keywords/feature-fusion","display_name":"feature fusion","score":0.7760000228881836},{"id":"https://openalex.org/keywords/cross-modal-attention","display_name":"cross-modal attention","score":0.9039999842643738},{"id":"https://openalex.org/keywords/pre-trained-language-models","display_name":"pre-trained language models","score":0.8130000233650208},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8629999756813049},{"id":"https://openalex.org/keywords/mortality-surveillance","display_name":"mortality surveillance","score":0.6980000138282776},{"id":"https://openalex.org/keywords/cause-specific-mortality-fractions","display_name":"cause-specific mortality fractions","score":0.765999972820282},{"id":"https://openalex.org/keywords/south-africa","display_name":"South Africa","score":0.8420000076293945},{"id":"https://openalex.org/keywords/benchmark-study","display_name":"benchmark study","score":0.4440000057220459},{"id":"https://openalex.org/keywords/calibration","display_name":"calibration","score":0.3529999852180481},{"id":"https://openalex.org/keywords/low-and-middle-income-countries","display_name":"low- and middle-income countries","score":0.7649999856948853}],"concepts":[{"id":"https://openalex.org/C2781410689","wikidata":"https://www.wikidata.org/wiki/Q17370247","display_name":"Verbal autopsy","level":4,"score":0.7056000232696533},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.590499997138977},{"id":"https://openalex.org/C199033989","wikidata":"https://www.wikidata.org/wiki/Q1318295","display_name":"Narrative","level":2,"score":0.5860999822616577},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5074999928474426},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.507099986076355},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5058000087738037},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.5045999884605408},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4738999903202057},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.4489000141620636},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.38830000162124634},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.3865000009536743},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.38109999895095825},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.37709999084472656},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.3725999891757965},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3709000051021576},{"id":"https://openalex.org/C195910791","wikidata":"https://www.wikidata.org/wiki/Q1324077","display_name":"Medical record","level":2,"score":0.34880000352859497},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.34279999136924744},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3077999949455261},{"id":"https://openalex.org/C138816342","wikidata":"https://www.wikidata.org/wiki/Q189603","display_name":"Public health","level":2,"score":0.29580000042915344},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.2872999906539917},{"id":"https://openalex.org/C3019952477","wikidata":"https://www.wikidata.org/wiki/Q1324077","display_name":"Health records","level":3,"score":0.2849999964237213},{"id":"https://openalex.org/C545542383","wikidata":"https://www.wikidata.org/wiki/Q2751242","display_name":"Medical emergency","level":1,"score":0.28290000557899475},{"id":"https://openalex.org/C3017944768","wikidata":"https://www.wikidata.org/wiki/Q1450463","display_name":"Poison control","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C2779473830","wikidata":"https://www.wikidata.org/wiki/Q1540899","display_name":"MEDLINE","level":2,"score":0.26809999346733093},{"id":"https://openalex.org/C2776361769","wikidata":"https://www.wikidata.org/wiki/Q821775","display_name":"Public health surveillance","level":3,"score":0.262800008058548},{"id":"https://openalex.org/C2779119184","wikidata":"https://www.wikidata.org/wiki/Q294350","display_name":"ALARM","level":2,"score":0.25220000743865967}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.64898/2026.10.15.26364476","doi":"https://doi.org/10.64898/2026.10.15.26364476","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.10.15.26364476","pdf_url":"https://www.medrxiv.org/content/medrxiv/early/2026/10/16/2026.10.15.26364476.full.pdf","source":{"id":"https://openalex.org/S3005729997","display_name":"medRxiv","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":["https://openalex.org/P4310315909"],"host_organization_lineage_names":["Cold Spring Harbor Laboratory Press"],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"medRxiv","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.64898/2026.10.15.26364476","doi":"https://doi.org/10.64898/2026.10.15.26364476","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.10.15.26364476","pdf_url":"https://www.medrxiv.org/content/medrxiv/early/2026/10/16/2026.10.15.26364476.full.pdf","source":{"id":"https://openalex.org/S3005729997","display_name":"medRxiv","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":["https://openalex.org/P4310315909"],"host_organization_lineage_names":["Cold Spring Harbor Laboratory Press"],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"medRxiv","raw_type":"posted-content"},"sustainable_development_goals":[{"display_name":"Good health and well-being","score":0.8371999859809875,"id":"https://metadata.un.org/sdg/3"}],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320309374","display_name":"Ohio State University","ror":"https://ror.org/00rs6vg23"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7222765853.pdf","grobid_xml":"https://content.openalex.org/works/W7222765853.grobid-xml"},"referenced_works_count":32,"referenced_works":["https://openalex.org/W158829242","https://openalex.org/W925979150","https://openalex.org/W1678356000","https://openalex.org/W1760646631","https://openalex.org/W2164561875","https://openalex.org/W2270210445","https://openalex.org/W2295598076","https://openalex.org/W2887172151","https://openalex.org/W2898227265","https://openalex.org/W2947667755","https://openalex.org/W2959425712","https://openalex.org/W2963341956","https://openalex.org/W2963716420","https://openalex.org/W2970161670","https://openalex.org/W2971258845","https://openalex.org/W3011307848","https://openalex.org/W3017961061","https://openalex.org/W3038294580","https://openalex.org/W3127206950","https://openalex.org/W3134239146","https://openalex.org/W3200899627","https://openalex.org/W4224980349","https://openalex.org/W4310568840","https://openalex.org/W4383677251","https://openalex.org/W4388199535","https://openalex.org/W4389238691","https://openalex.org/W4398141723","https://openalex.org/W4402517573","https://openalex.org/W7081579838","https://openalex.org/W7117138703","https://openalex.org/W7118388384","https://openalex.org/W7214789911"],"related_works":["https://openalex.org/W7078150166","https://openalex.org/W4297313072","https://openalex.org/W7081579838","https://openalex.org/W4383677251","https://openalex.org/W3038294580","https://openalex.org/W7117138703","https://openalex.org/W3212470077","https://openalex.org/W7213956671","https://openalex.org/W7223578085","https://openalex.org/W2959425712"],"abstract_inverted_index":{"Abstract":[0],"Background":[1],"Timely":[2],"cause-of-death":[3],"(COD)":[4],"information":[5],"guides":[6],"health":[7],"policy,":[8],"yet":[9],"many":[10],"deaths":[11,27,99,112,256],"in":[12,113,205,230,235,248,304,308,329,335,445],"low-":[13],"and":[14,37,127,150,158,176,201,233,243,268,284,332,372,411,416,431,480,493],"middle-income":[15],"countries":[16],"are":[17],"not":[18,73],"medically":[19],"certified.":[20],"In":[21],"Africa,":[22],"fewer":[23],"than":[24,211,397],"10%":[25],"of":[26,110,217,255,378],"are.":[28],"Verbal":[29],"autopsy":[30],"(VA)":[31],"yields":[32],"a":[33,38,107,164,473],"structured":[34,62,147,244,436],"symptom":[35],"questionnaire,":[36,49],"free-text":[39],"narrative.":[40],"Mortality":[41],"surveillance":[42],"increasingly":[43],"records":[44,97],"the":[45,48,51,61,121,131,151,155,173,187,265,272,276,285,292,295,309,320,341,349,374,383,427,463,486],"narrative":[46,242,266,277],"alongside":[47],"but":[50,71,250],"algorithms":[52],"used":[53],"to":[54,75,88,300,357,476],"produce":[55],"VA-based":[56],"statistics":[57],"were":[58,183,261],"designed":[59],"for":[60,98,281,288,485],"responses":[63,148],"only,":[64],"so":[65],"narratives":[66,140,422,471],"inform":[67],"physician":[68],"COD":[69,77,90,447],"assignment":[70],"do":[72],"contribute":[74],"algorithmic":[76],"assignment.":[78,91,448],"We":[79,93,124,161,194],"evaluated":[80,195],"whether":[81],"supervised":[82,221,384,459],"models":[83,136,145],"could":[84],"use":[85],"both":[86],"modalities":[87],"improve":[89],"Methods":[92],"analyzed":[94],"4,999":[95],"VA":[96,478],"at":[100,154],"ages":[101],"12":[102],"years":[103],"or":[104],"older":[105],"from":[106,264,271,348,354],"national":[108],"sample":[109],"registered":[111],"South":[114],"Africa":[115],"(RSA),":[116],"using":[117,421,435],"physician-assigned":[118],"CODs":[119],"as":[120,472],"reference":[122],"standard.":[123],"trained":[125],"unimodal":[126,175,293,351],"multimodal":[128,177,440],"pipelines":[129,213,245,297,385,391,434,460],"on":[130,139,146,172,214,224,369,426,455],"labelled":[132],"records:":[133],"pretrained":[134],"language":[135,169],"(PLMs)":[137],"fine-tuned":[138],"alone,":[141,149],"machine":[142],"learning":[143],"(ML)":[144],"two":[152],"combined":[153],"data,":[156],"feature":[157,314],"decision":[159],"levels.":[160],"also":[162],"prompted":[163],"locally":[165],"deployed":[166],"generative":[167],"large":[168],"model":[170],"(LLM)":[171],"same":[174],"inputs,":[178],"without":[179],"parameter":[180],"updates.":[181],"All":[182],"compared":[184],"with":[185,275,316,360],"InSilicoVA,":[186],"operational":[188],"comparator":[189],"representing":[190],"current":[191,428],"automated":[192,477],"practice.":[193],"individual-level":[196,370],"assignment,":[197],"population-level":[198],"agreement,":[199],"calibration,":[200],"generalization":[202],"across":[203,402,491],"provinces":[204],"RSA,":[206],"reporting":[207],"these":[208],"metrics":[209,371],"rather":[210],"ranking":[212],"any":[215,379],"one":[216],"them.":[218],"Results":[219],"Every":[220],"pipeline":[222,267,278,287,380],"improved":[223,319,425],"InSilicoVA":[225,368],"by":[226,252],"8.5\u201315.9":[227],"percentage":[228,302],"points":[229,303,328],"top-1":[231,305,409],"accuracy":[232,331,344,377],"3.5\u20137.1":[234],"cause-specific":[236],"mortality":[237],"fraction":[238],"(CSMF)":[239],"accuracy.":[240,389],"The":[241,365,449],"performed":[246],"similarly":[247],"aggregate":[249],"differed":[251],"cause:":[253],"63.9%":[254],"received":[257],"identical":[258],"predictions,":[259],"11.6%":[260],"correct":[262],"only":[263,270],"13.6%":[269],"questionnaire":[273,286],"pipeline,":[274],"more":[279],"accurate":[280],"noncommunicable":[282],"causes":[283,323],"communicable":[289],"causes.":[290],"Against":[291],"approaches,":[294],"decision-fusion":[296],"added":[298],"4.2":[299],"5.4":[301],"accuracy,":[306],"concentrated":[307],"most":[310],"common":[311,322],"causes,":[312,457],"whereas":[313,458],"fusion":[315,339,483],"cross-modal":[317],"attention":[318],"less":[321],"instead,":[324],"adding":[325],"6.6":[326],"(1.7\u201311.5)":[327],"balanced":[330,376],"5.5":[333],"(0.6\u201310.3)":[334],"macro":[336,414],"F1.":[337],"Feature":[338],"reached":[340,373],"highest":[342,375],"CSMF":[343,388,417],"(95.7%).":[345],"Paired":[346],"differences":[347],"better":[350,461],"representative":[352,482],"ranged":[353],"\u2212":[355],"0.2":[356],"+3.4":[358],"points,":[359],"all":[361,403],"intervals":[362],"including":[363,407],"zero.":[364],"LLM":[366,451],"exceeded":[367],"(60.6%),":[381],"while":[382,438],"retained":[386],"higher":[387],"Supervised":[390,399],"had":[392],"lower":[393],"expected":[394],"calibration":[395],"error":[396],"InSilicoVA.":[398],"gains":[400,444],"held":[401],"nine":[404],"held-out":[405],"provinces,":[406],"every":[408,413],"comparison":[410],"nearly":[412],"F1":[415],"comparison.":[418],"Conclusions":[419],"PLMs":[420],"alone":[423],"substantially":[424],"best":[429],"practice":[430],"matched":[432],"ML":[433],"responses,":[437],"multiple":[439],"strategies":[441,484],"produced":[442],"further":[443],"individual":[446],"zero-shot":[450],"showed":[452],"particular":[453],"strength":[454],"rare":[456],"recovered":[462],"population":[464],"cause":[465],"distribution.":[466],"These":[467],"findings":[468],"support":[469,497],"incorporating":[470],"core":[474],"input":[475],"systems":[479],"evaluating":[481],"intended":[487],"output.":[488],"External":[489],"validation":[490],"settings":[492],"age":[494],"groups":[495],"will":[496],"broader":[498],"implementation.":[499]},"counts_by_year":[],"updated_date":"2026-11-09T13:22:03.289127","created_date":"2026-10-17T00:00:00"}