{"id":"https://openalex.org/W7134847464","doi":"https://doi.org/10.48550/arxiv.2603.07443","title":"Med-Evo: Test-time Self-evolution for Medical Multimodal Large Language Models","display_name":"Med-Evo: Test-time Self-evolution for Medical Multimodal Large Language Models","publication_year":2026,"publication_date":"2026-03-08","ids":{"openalex":"https://openalex.org/W7134847464","doi":"https://doi.org/10.48550/arxiv.2603.07443"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2603.07443","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101204537","display_name":"Dunyuan Xu","orcid":"https://orcid.org/0000-0001-7600-9384"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Dunyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046094203","display_name":"Xikai Yang","orcid":"https://orcid.org/0000-0003-1762-9684"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Xikai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100786224","display_name":"Juzheng Miao","orcid":"https://orcid.org/0000-0002-7011-1481"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miao, Juzheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070831420","display_name":"Yaoqian Li","orcid":"https://orcid.org/0000-0002-5189-8151"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yaoqian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128687437","display_name":"Jinpeng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jinpeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128630602","display_name":"Pheng-Ann Heng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heng, Pheng-Ann","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":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.2918539,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.4372999966144562,"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":0.4372999966144562,"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":0.19280000030994415,"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":0.11879999935626984,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.5727999806404114},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5534999966621399},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.5527999997138977},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5037000179290771},{"id":"https://openalex.org/keywords/precision-and-recall","display_name":"Precision and recall","score":0.4041000008583069},{"id":"https://openalex.org/keywords/semantic-similarity","display_name":"Semantic similarity","score":0.38280001282691956},{"id":"https://openalex.org/keywords/unified-medical-language-system","display_name":"Unified Medical Language System","score":0.3756999969482422},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.36890000104904175},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.36480000615119934}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7874000072479248},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6320000290870667},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5727999806404114},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5534999966621399},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.5527999997138977},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5406000018119812},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5037000179290771},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.4041000008583069},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.38280001282691956},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37869998812675476},{"id":"https://openalex.org/C69505689","wikidata":"https://www.wikidata.org/wiki/Q455338","display_name":"Unified Medical Language System","level":2,"score":0.3756999969482422},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.36890000104904175},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.36480000615119934},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.35359999537467957},{"id":"https://openalex.org/C144986985","wikidata":"https://www.wikidata.org/wiki/Q871236","display_name":"Hierarchical database model","level":2,"score":0.34049999713897705},{"id":"https://openalex.org/C100660578","wikidata":"https://www.wikidata.org/wiki/Q18733","display_name":"Recall","level":2,"score":0.3375999927520752},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.3370000123977661},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3343000113964081},{"id":"https://openalex.org/C16910744","wikidata":"https://www.wikidata.org/wiki/Q7705759","display_name":"Test data","level":2,"score":0.33410000801086426},{"id":"https://openalex.org/C146599234","wikidata":"https://www.wikidata.org/wiki/Q511093","display_name":"Centroid","level":2,"score":0.3165999948978424},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C90312973","wikidata":"https://www.wikidata.org/wiki/Q7449052","display_name":"Semantic data model","level":2,"score":0.2840000092983246},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.28380000591278076},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.2750999927520752},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2712000012397766},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.26109999418258667},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.25679999589920044}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2603.07443","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2603.07443","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.07443","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":"pmh:doi:10.48550/arxiv.2603.07443","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Medical":[0],"Multimodal":[1],"Large":[2],"Language":[3],"Models":[4],"(MLLMs)":[5],"have":[6],"demonstrated":[7],"remarkable":[8],"capabilities":[9],"across":[10],"diverse":[11],"healthcare":[12],"tasks.":[13],"However,":[14],"current":[15],"post-training":[16],"strategies,":[17],"such":[18],"as":[19],"supervised":[20],"fine-tuning":[21],"and":[22,64,81,140,152,165,184],"reinforcement":[23,102],"learning,":[24],"heavily":[25],"depend":[26],"on":[27,160,187],"substantial":[28],"annotated":[29],"data":[30,38,55,62,70],"while":[31],"overlooking":[32],"the":[33,60,92,188,194],"potential":[34],"of":[35,172,181,196],"unlabeled":[36,79],"test":[37,69],"for":[39,96],"model":[40,106],"enhancement.":[41],"This":[42],"limitation":[43],"becomes":[44],"particularly":[45],"pronounced":[46],"in":[47,73,137],"medical":[48,54,97,162],"domains,":[49],"where":[50],"acquiring":[51],"extensive":[52],"labeled":[53,111],"is":[56],"difficult":[57],"due":[58],"to":[59,104,133,155],"strict":[61],"sensitivity":[63],"annotation":[65],"complexity.":[66],"Moreover,":[67],"leveraging":[68],"poses":[71],"challenges":[72],"generating":[74],"reliable":[75],"supervision":[76],"signals":[77],"from":[78,128],"samples":[80],"maintaining":[82],"stable":[83],"self-evolution.":[84],"To":[85],"address":[86],"these":[87],"limitations,":[88],"we":[89],"propose":[90],"Med-Evo,":[91],"first":[93],"self-evolution":[94],"framework":[95,114],"MLLMs":[98,168],"that":[99,124,145],"utilizes":[100],"label-free":[101],"learning":[103],"promote":[105],"performance":[107],"without":[108],"requiring":[109],"additional":[110],"data.":[112],"Our":[113],"introduces":[115],"two":[116,166],"key":[117],"innovations:":[118],"$1)$":[119],"Feature-driven":[120],"Pseudo":[121],"Labeling":[122],"(FPL)":[123],"identifies":[125],"semantic":[126,153],"centroids":[127],"all":[129],"heterogeneous":[130],"candidate":[131],"responses":[132],"select":[134],"pseudo":[135],"labels":[136],"each":[138],"rollout,":[139],"$2)$":[141],"Hard-Soft":[142],"Reward":[143],"(HSR)":[144],"combines":[146],"exact":[147],"match":[148],"with":[149,178],"token-level":[150],"assessment":[151],"similarity":[154],"provide":[156],"hierarchical":[157],"reward.":[158],"Experiments":[159],"three":[161],"VQA":[163],"benchmarks":[164],"base":[167],"show":[169],"clear":[170],"advantages":[171],"our":[173,197],"approach":[174],"over":[175],"SOTA":[176],"methods,":[177],"significant":[179],"improvements":[180],"10.43\\%":[182],"accuracy":[183],"4.68\\%":[185],"recall":[186],"SLAKE":[189],"dataset":[190],"using":[191],"Qwen2.5-VL,":[192],"showing":[193],"effectiveness":[195],"method.":[198]},"counts_by_year":[],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2026-03-11T00:00:00"}