{"id":"https://openalex.org/W7146975714","doi":"https://doi.org/10.1109/cw68232.2025.00066","title":"Dual Fine-Tuning and Dual Alignment Training for Medical Language Models","display_name":"Dual Fine-Tuning and Dual Alignment Training for Medical Language Models","publication_year":2025,"publication_date":"2025-10-14","ids":{"openalex":"https://openalex.org/W7146975714","doi":"https://doi.org/10.1109/cw68232.2025.00066"},"language":null,"primary_location":{"id":"doi:10.1109/cw68232.2025.00066","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cw68232.2025.00066","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Conference on Cyberworlds (CW\uff09","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031751659","display_name":"Junyu Lu","orcid":"https://orcid.org/0009-0008-0574-9519"},"institutions":[{"id":"https://openalex.org/I4210165204","display_name":"Zhuhai Institute of Advanced Technology","ror":"https://ror.org/05r1mzq61","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210145761","https://openalex.org/I4210165204"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Junyu Lu","raw_affiliation_strings":["Beijing Institute of Technology,Aerospace Information Domain,Zhuhai,China"],"affiliations":[{"raw_affiliation_string":"Beijing Institute of Technology,Aerospace Information Domain,Zhuhai,China","institution_ids":["https://openalex.org/I4210165204"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5132547388","display_name":"Zehua Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210095683","display_name":"Jinzhong University","ror":"https://ror.org/007ywhm20","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210095683"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zehua Chen","raw_affiliation_strings":["College of Artificial Intelligence, Taiyuan University of Technology,Jinzhong,China"],"affiliations":[{"raw_affiliation_string":"College of Artificial Intelligence, Taiyuan University of Technology,Jinzhong,China","institution_ids":["https://openalex.org/I4210095683"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5031751659"],"corresponding_institution_ids":["https://openalex.org/I4210165204"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.87487981,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"406","last_page":"413"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.4230000078678131,"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":0.4230000078678131,"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":0.17659999430179596,"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/T10028","display_name":"Topic Modeling","score":0.1559000015258789,"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/dual","display_name":"Dual (grammatical number)","score":0.7197999954223633},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.6395000219345093},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5375000238418579},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5056999921798706},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.49779999256134033},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.47870001196861267},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4415999948978424},{"id":"https://openalex.org/keywords/co-training","display_name":"Co-training","score":0.43470001220703125}],"concepts":[{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.7197999954223633},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7063000202178955},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6395000219345093},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5375000238418579},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5321000218391418},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5056999921798706},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.49779999256134033},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.47870001196861267},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46070000529289246},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4415999948978424},{"id":"https://openalex.org/C2776959682","wikidata":"https://www.wikidata.org/wiki/Q17005296","display_name":"Co-training","level":3,"score":0.43470001220703125},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.40689998865127563},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3953000009059906},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3199999928474426},{"id":"https://openalex.org/C20556612","wikidata":"https://www.wikidata.org/wiki/Q4469374","display_name":"Volume (thermodynamics)","level":2,"score":0.3043000102043152},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.2962999939918518},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2825999855995178},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.2653999924659729},{"id":"https://openalex.org/C69505689","wikidata":"https://www.wikidata.org/wiki/Q455338","display_name":"Unified Medical Language System","level":2,"score":0.2605000138282776},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.25699999928474426},{"id":"https://openalex.org/C2780813799","wikidata":"https://www.wikidata.org/wiki/Q3274237","display_name":"Zero (linguistics)","level":2,"score":0.2540000081062317}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cw68232.2025.00066","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cw68232.2025.00066","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Conference on Cyberworlds (CW\uff09","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W2763110165","https://openalex.org/W2911489562","https://openalex.org/W3046375318","https://openalex.org/W3100355250","https://openalex.org/W3129831491","https://openalex.org/W3176580738","https://openalex.org/W3205803342","https://openalex.org/W4384071683","https://openalex.org/W4385245566","https://openalex.org/W4392240262","https://openalex.org/W4401386758","https://openalex.org/W4415798212","https://openalex.org/W7133208539","https://openalex.org/W7133211372","https://openalex.org/W7133227460","https://openalex.org/W7133233472"],"related_works":[],"abstract_inverted_index":{"The":[0,59],"vast":[1],"volume":[2],"of":[3,83,103],"medical":[4,51,98,126],"information":[5],"in":[6,119,149],"the":[7,56,64,79,93,97,113,134],"real":[8],"world":[9],"demands":[10],"intelligent":[11],"assistant":[12],"systems":[13],"for":[14],"effective":[15],"integration":[16],"and":[17,34,37,44,75,107,124,151],"summarization.":[18],"In":[19],"response":[20],"to":[21,48],"this,":[22],"we":[23],"designed":[24],"a":[25,50],"fourstage":[26],"training":[27,40,61,95],"pipeline":[28],"comprising":[29],"dual":[30,38,131],"fine-tuning":[31],"(continued":[32],"pre-training":[33],"supervised":[35],"fine-tuning)":[36],"alignment":[39,132],"(Direct":[41],"Preference":[42],"Optimization":[43],"Proximal":[45],"Policy":[46],"Optimization),":[47],"build":[49],"language":[52,85,99],"model":[53,86,100],"based":[54],"on":[55,105,109],"open-source":[57],"LLaMA-3.1-8B.":[58],"entire":[60],"process":[62],"adopts":[63],"DeepSpeed":[65],"framework,":[66],"combined":[67],"with":[68],"ZeRO-1":[69],"optimization,":[70],"Quantized":[71],"Low-Rank":[72],"Adaptation":[73],"(QLoRA),":[74],"Flash-Attention":[76],"2,":[77],"alleviating":[78],"substantial":[80],"computational":[81],"costs":[82],"large":[84],"(LLM)":[87],"training.":[88],"Results":[89],"show":[90],"that":[91],"after":[92,130],"four-stage":[94],"process,":[96],"achieved":[101],"scores":[102],"50.49":[104],"CMMLU":[106],"35.00":[108],"CMB,":[110],"both":[111,120],"outperforming":[112],"baseline":[114],"LLaMA-3.1-8B-Instruct,":[115],"indicating":[116],"significant":[117],"improvements":[118],"general":[121],"Chinese":[122,125],"capabilities":[123],"domain":[127],"capabilities.":[128],"Furthermore,":[129],"training,":[133],"model's":[135],"reward":[136],"score":[137],"increased":[138],"by":[139,144],"30%,":[140],"while":[141],"toxicity":[142],"decreased":[143],"25%,":[145],"demonstrating":[146],"considerable":[147],"enhancements":[148],"safety":[150],"harmlessness.":[152]},"counts_by_year":[],"updated_date":"2026-04-02T13:53:19.096889","created_date":"2026-04-02T00:00:00"}