{"id":"https://openalex.org/W7165118072","doi":"https://doi.org/10.48550/arxiv.2606.19266","title":"Trade-offs in Medical LLM Adaptation: An Empirical Study in French QA","display_name":"Trade-offs in Medical LLM Adaptation: An Empirical Study in French QA","publication_year":2026,"publication_date":"2026-06-17","ids":{"openalex":"https://openalex.org/W7165118072","doi":"https://doi.org/10.48550/arxiv.2606.19266"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.19266","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.19266","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.2606.19266","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138896310","display_name":"Ikram Belmadani","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Belmadani, Ikram","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132861601","display_name":"Oumaima El Khettari","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Khettari, Oumaima El","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135118672","display_name":"Carlos Ramisch","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ramisch, Carlos","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110498175","display_name":"Fr\u00e9d\u00e9ric B\u00e9chet","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bechet, Frederic","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138907051","display_name":"Richard Dufour","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dufour, Richard","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138905540","display_name":"Benoit Favre","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Favre, Benoit","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/T10028","display_name":"Topic Modeling","score":0.6984999775886536,"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":0.6984999775886536,"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/T13629","display_name":"Text Readability and Simplification","score":0.10010000318288803,"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":0.06610000133514404,"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/adaptation","display_name":"Adaptation (eye)","score":0.7067999839782715},{"id":"https://openalex.org/keywords/domain-adaptation","display_name":"Domain adaptation","score":0.6406999826431274},{"id":"https://openalex.org/keywords/empirical-research","display_name":"Empirical research","score":0.5493999719619751},{"id":"https://openalex.org/keywords/initialization","display_name":"Initialization","score":0.5371000170707703},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.4424999952316284},{"id":"https://openalex.org/keywords/contrast","display_name":"Contrast (vision)","score":0.34869998693466187}],"concepts":[{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.7067999839782715},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6855999827384949},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.6406999826431274},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.5493999719619751},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.5371000170707703},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.48829999566078186},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.4424999952316284},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4198000133037567},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.34869998693466187},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.3440999984741211},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.3206999897956848},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2992999851703644},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.2766999900341034},{"id":"https://openalex.org/C534262118","wikidata":"https://www.wikidata.org/wiki/Q177719","display_name":"Medical diagnosis","level":2,"score":0.25999999046325684}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.19266","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.19266","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.2606.19266","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.19266","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":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.5177342891693115}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"development":[1],"of":[2,24,34],"large":[3],"language":[4],"models":[5],"(LLMs)":[6],"has":[7],"led":[8],"to":[9,16,154],"an":[10],"increased":[11],"focus":[12],"on":[13],"their":[14,56],"adaptation":[15,26,37,70,153,164],"specialized":[17],"domains":[18],"and":[19,55,64,81,87,93,111,120,138],"languages,":[20],"yet":[21],"the":[22,102],"effectiveness":[23],"domain":[25,36],"strategies":[27,165],"remains":[28],"unclear.":[29],"We":[30,47,76],"present":[31],"a":[32,44,118],"study":[33],"medical":[35,40],"using":[38,90],"French":[39,152],"question-answering":[41],"(QA)":[42],"as":[43],"case":[45],"study.":[46],"compare":[48],"continual":[49],"pretraining":[50],"(CPT),":[51],"supervised":[52],"fine-tuning":[53],"(SFT),":[54],"combination":[57],"across":[58],"three":[59,65],"model":[60,74],"families,":[61],"multiple":[62],"sizes,":[63],"initialization":[66],"types,":[67],"explicitly":[68],"disentangling":[69],"effects":[71],"from":[72,151],"base":[73],"choice.":[75],"evaluate":[77],"both":[78],"multiple-choice":[79],"(MCQA)":[80],"open-ended":[82],"QA":[83],"(OEQA)":[84],"under":[85,166],"greedy":[86],"constrained":[88],"decoding":[89],"automatic":[91],"metrics":[92],"LLM-as-a-Judge":[94],"evaluation.":[95,144],"For":[96,123],"MCQA,":[97],"CPT+SFT":[98,139],"most":[99],"often":[100,132],"achieves":[101],"best":[103],"scores,":[104],"but":[105],"gains":[106],"over":[107],"SFT":[108,117,131],"are":[109,140],"small":[110],"frequently":[112],"not":[113],"statistically":[114],"significant,":[115],"making":[116],"strong":[119],"cost-effective":[121],"default.":[122],"OEQA,":[124],"CPT":[125],"consistently":[126],"improves":[127],"overlap-based":[128],"metrics,":[129],"while":[130],"degrades":[133],"generation":[134],"quality;":[135],"instruction":[136],"tuning":[137],"preferred":[141],"by":[142],"LLM-based":[143],"Cross-lingual":[145],"experiments":[146],"further":[147],"show":[148],"effective":[149],"transfer":[150],"English":[155],"benchmarks.":[156],"Overall,":[157],"we":[158],"provide":[159],"practical":[160],"guidelines":[161],"for":[162],"selecting":[163],"computational":[167],"constraints.":[168]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-19T00:00:00"}