{"id":"https://openalex.org/W7164805222","doi":"https://doi.org/10.48550/arxiv.2606.13940","title":"Can Post-Training Turn LLMs into Good Medical Coders? An Empirical Study of Generative ICD Coding","display_name":"Can Post-Training Turn LLMs into Good Medical Coders? An Empirical Study of Generative ICD Coding","publication_year":2026,"publication_date":"2026-06-11","ids":{"openalex":"https://openalex.org/W7164805222","doi":"https://doi.org/10.48550/arxiv.2606.13940"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.13940","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.13940","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2606.13940","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138677497","display_name":"Ziqing Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Ziqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138679046","display_name":"Weihao Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Weihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138670130","display_name":"Shijie Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Shijie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138672986","display_name":"Yuan Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138683888","display_name":"Kaize Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Kaize","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/T13702","display_name":"Machine Learning in Healthcare","score":0.6628000140190125,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.6628000140190125,"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/T14400","display_name":"Medical Coding and Health Information","score":0.15119999647140503,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.036400001496076584,"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/generative-grammar","display_name":"Generative grammar","score":0.7928000092506409},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.6499999761581421},{"id":"https://openalex.org/keywords/coding","display_name":"Coding (social sciences)","score":0.5396999716758728},{"id":"https://openalex.org/keywords/empirical-research","display_name":"Empirical research","score":0.5361999869346619},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.49630001187324524},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.4771000146865845},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.44830000400543213}],"concepts":[{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.7928000092506409},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.6499999761581421},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.5396999716758728},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.5361999869346619},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5291000008583069},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4975000023841858},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.49630001187324524},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4797999858856201},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4771000146865845},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.44830000400543213},{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.36320000886917114},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.3151000142097473},{"id":"https://openalex.org/C154874363","wikidata":"https://www.wikidata.org/wiki/Q3518464","display_name":"Medical classification","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3061999976634979},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.2881999909877777},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.27399998903274536},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C138816342","wikidata":"https://www.wikidata.org/wiki/Q189603","display_name":"Public health","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.13940","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.13940","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2606.13940","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.13940","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":[{"id":"https://metadata.un.org/sdg/10","score":0.7029517889022827,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Automated":[0],"International":[1],"Classification":[2],"of":[3,50,60,129],"Diseases":[4],"(ICD)":[5],"coding":[6],"is":[7,90,162,172],"a":[8,56,80,109],"core":[9],"medical-coding":[10],"task":[11],"for":[12,62,98,131,176],"billing,":[13],"epidemiology,":[14],"and":[15,76,83,147,174,185],"clinical":[16],"decision":[17],"support.":[18],"Generative":[19],"large":[20],"language":[21],"models":[22],"(LLMs)":[23],"are":[24],"often":[25],"reported":[26],"as":[27,40],"weak":[28],"medical":[29],"coders,":[30],"but":[31,168],"this":[32,89],"finding":[33],"mainly":[34],"comes":[35],"from":[36],"inference-time":[37],"settings":[38],"such":[39],"prompting,":[41,73],"retrieval,":[42],"reranking,":[43],"or":[44],"tool":[45],"use,":[46],"leaving":[47],"the":[48,91,127,136,159,164,170],"role":[49],"task-specific":[51],"post-training":[52,61,97],"underexplored.":[53],"We":[54,105,179],"present":[55],"controlled":[57],"empirical":[58],"study":[59,93],"generative":[63,99,165],"ICD":[64,103,132],"coding,":[65],"comparing":[66],"discriminative":[67],"baselines":[68],"with":[69],"LLM":[70,100],"coders":[71,101],"across":[72],"supervised":[74],"fine-tuning,":[75],"reinforcement":[77],"learning":[78],"under":[79],"common":[81],"protocol":[82],"metric":[84],"set.":[85],"To":[86],"our":[87,181],"knowledge,":[88],"first":[92],"to":[94,115],"evaluate":[95],"RL-based":[96],"in":[102],"coding.":[104,133],"further":[106,141],"introduce":[107],"PHI,":[108],"diagnostic":[110],"curriculum":[111],"that":[112,122,158],"extends":[113],"GRPO":[114,140],"refine":[116],"missed-code":[117],"cases.":[118],"Our":[119],"results":[120],"show":[121],"prompting-only":[123],"evaluation":[124],"substantially":[125],"underestimates":[126],"potential":[128],"LLMs":[130],"SFT":[134],"provides":[135,149],"main":[137,160],"capability":[138],"jump,":[139],"improves":[142],"code-set":[143],"prediction":[144],"beyond":[145],"SFT,":[146],"PHI":[148],"targeted":[150],"gains":[151],"on":[152],"macro-level":[153],"performance.":[154],"These":[155],"findings":[156],"suggest":[157],"bottleneck":[161],"not":[163],"formulation":[166],"alone,":[167],"how":[169],"model":[171],"adapted":[173],"optimized":[175],"full-taxonomy":[177],"recall.":[178],"release":[180],"code,":[182],"data":[183],"splits,":[184],"checkpoints":[186],"at":[187],"https://github.com/AlexandreWANG915/LLM4ICD.":[188]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-16T00:00:00"}