{"id":"https://openalex.org/W7204995493","doi":"https://doi.org/10.21203/rs.3.rs-10675861/v1","title":"Conservative Proxy Prompting for Reliable Frozen Vision\u2013Language Models with Missing Modalities","display_name":"Conservative Proxy Prompting for Reliable Frozen Vision\u2013Language Models with Missing Modalities","publication_year":2026,"publication_date":"2026-09-02","ids":{"openalex":"https://openalex.org/W7204995493","doi":"https://doi.org/10.21203/rs.3.rs-10675861/v1"},"language":"en","primary_location":{"id":"doi:10.21203/rs.3.rs-10675861/v1","doi":"https://doi.org/10.21203/rs.3.rs-10675861/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-10675861/v1","pdf_url":"https://www.researchsquare.com/article/rs-10675861/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"volume":null,"type":"preprint","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.researchsquare.com/article/rs-10675861/latest.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5045778370","display_name":"Xiaoyong Mei","orcid":"https://orcid.org/0000-0001-6422-1724","observed_orcids":["https://orcid.org/0000-0001-6422-1724"]},"institutions":[{"id":"https://openalex.org/I135237710","display_name":"Zhejiang Normal University","ror":"https://ror.org/01vevwk45","country_code":"CN","type":"education","lineage":["https://openalex.org/I135237710"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoyong Mei","raw_affiliation_strings":["Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University","institution_ids":["https://openalex.org/I135237710"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148949781","display_name":"Wei Ji","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I135237710","display_name":"Zhejiang Normal University","ror":"https://ror.org/01vevwk45","country_code":"CN","type":"education","lineage":["https://openalex.org/I135237710"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Ji","raw_affiliation_strings":["Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University","institution_ids":["https://openalex.org/I135237710"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126136235","display_name":"Dong Jiale","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I135237710","display_name":"Zhejiang Normal University","ror":"https://ror.org/01vevwk45","country_code":"CN","type":"education","lineage":["https://openalex.org/I135237710"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiale Dong","raw_affiliation_strings":["Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University","institution_ids":["https://openalex.org/I135237710"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013607751","display_name":"Chaonan Duan","orcid":"https://orcid.org/0009-0002-9315-5417","observed_orcids":["https://orcid.org/0009-0002-9315-5417"]},"institutions":[{"id":"https://openalex.org/I135237710","display_name":"Zhejiang Normal University","ror":"https://ror.org/01vevwk45","country_code":"CN","type":"education","lineage":["https://openalex.org/I135237710"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chao Duan","raw_affiliation_strings":["Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University","institution_ids":["https://openalex.org/I135237710"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101868777","display_name":"Mingyan Zhang","orcid":"https://orcid.org/0000-0002-7602-504X","observed_orcids":["https://orcid.org/0000-0002-7602-504X"]},"institutions":[{"id":"https://openalex.org/I135237710","display_name":"Zhejiang Normal University","ror":"https://ror.org/01vevwk45","country_code":"CN","type":"education","lineage":["https://openalex.org/I135237710"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mingyan Zhang","raw_affiliation_strings":["Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang Key Laboratory of Intelligent Education Technology and Application, Zhejiang Normal University","institution_ids":["https://openalex.org/I135237710"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024126710","display_name":"Fudan Zheng","orcid":"https://orcid.org/0000-0002-9664-012X","observed_orcids":["https://orcid.org/0000-0002-9664-012X"]},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fudan Zheng","raw_affiliation_strings":["Sun Yat-Sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sun Yat-Sen University","institution_ids":["https://openalex.org/I157773358"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/missing-modalities","display_name":"missing modalities","score":0.8569999933242798},{"id":"https://openalex.org/keywords/prompt-learning","display_name":"prompt learning","score":0.6790000200271606},{"id":"https://openalex.org/keywords/multimodal-fusion","display_name":"multimodal fusion","score":0.8560000061988831},{"id":"https://openalex.org/keywords/incomplete-observations","display_name":"incomplete observations","score":0.6850000023841858},{"id":"https://openalex.org/keywords/parameter-efficient-fine-tuning","display_name":"parameter-efficient fine-tuning","score":0.7739999890327454},{"id":"https://openalex.org/keywords/multimodal-transfer","display_name":"multimodal transfer","score":0.34299999475479126}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7232999801635742},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.699400007724762},{"id":"https://openalex.org/C2780148112","wikidata":"https://www.wikidata.org/wiki/Q1432581","display_name":"Proxy (statistics)","level":2,"score":0.6869000196456909},{"id":"https://openalex.org/C9357733","wikidata":"https://www.wikidata.org/wiki/Q6878417","display_name":"Missing data","level":2,"score":0.6324999928474426},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5194000005722046},{"id":"https://openalex.org/C61272859","wikidata":"https://www.wikidata.org/wiki/Q7834031","display_name":"Transferability","level":3,"score":0.5055000185966492},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4837000072002411},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.45339998602867126},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.44690001010894775},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.42340001463890076},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.37929999828338623},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3248000144958496},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.29660001397132874},{"id":"https://openalex.org/C27158222","wikidata":"https://www.wikidata.org/wiki/Q5532422","display_name":"Generalizability theory","level":2,"score":0.29120001196861267}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21203/rs.3.rs-10675861/v1","doi":"https://doi.org/10.21203/rs.3.rs-10675861/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-10675861/v1","pdf_url":"https://www.researchsquare.com/article/rs-10675861/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.21203/rs.3.rs-10675861/v1","doi":"https://doi.org/10.21203/rs.3.rs-10675861/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-10675861/v1","pdf_url":"https://www.researchsquare.com/article/rs-10675861/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"display_name":"Zero hunger","id":"https://metadata.un.org/sdg/2","score":0.48818233609199524}],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320327281","display_name":"Jilin Office of Philosophy and Social Science","ror":"https://ror.org/02y7tmn58"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204995493.pdf","grobid_xml":"https://content.openalex.org/works/W7204995493.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4399423378","https://openalex.org/W4403594035","https://openalex.org/W4416347319","https://openalex.org/W7140863123","https://openalex.org/W7127572139","https://openalex.org/W7161144574","https://openalex.org/W7207669827","https://openalex.org/W4415797272","https://openalex.org/W4406975831","https://openalex.org/W4414360615"],"abstract_inverted_index":{"Abstract":[0],"Frozen":[1],"vision--language":[2,60],"models":[3,61],"have":[4],"achieved":[5],"strong":[6],"transferability":[7],"by":[8],"learning":[9,75],"aligned":[10],"image--text":[11],"representations,":[12],"but":[13],"their":[14,92],"reliability":[15,71],"can":[16],"degrade":[17],"when":[18],"one":[19],"modality":[20],"becomes":[21],"unavailable":[22,31],"in":[23],"practical":[24],"scenarios.":[25],"Existing":[26],"missing-modality":[27],"methods":[28],"often":[29],"reconstruct":[30],"information;":[32],"however,":[33],"recovered":[34],"representations":[35,85],"may":[36],"contain":[37],"uncertain":[38],"signals":[39],"and":[40,76,89,101,107,144,151],"introduce":[41],"misleading":[42],"evidence.":[43],"To":[44],"address":[45],"this":[46],"issue,":[47],"we":[48],"propose":[49],"Conservative":[50],"Proxy":[51],"Prompting":[52],"(CPP),":[53],"a":[54,130],"parameter-efficient":[55],"framework":[56],"for":[57,86],"adapting":[58],"frozen":[59],"under":[62,110],"incomplete":[63,124],"observations.":[64],"CPP":[65,81,118],"separates":[66],"task":[67],"adaptation":[68],"from":[69],"proxy":[70,78,84],"through":[72],"lightweight":[73],"prompt":[74],"reliability-aware":[77],"utilization.":[79],"Specifically,":[80],"constructs":[82],"feature-space":[83],"missing":[87,112],"modalities":[88],"conservatively":[90],"regulates":[91],"contribution":[93],"during":[94],"multimodal":[95,125],"fusion.":[96],"Experiments":[97],"on":[98],"MM-IMDb,":[99],"Food-101,":[100],"Hateful":[102],"Memes":[103],"evaluate":[104],"text-missing,":[105],"image-missing,":[106],"mixed-missing":[108],"settings":[109],"different":[111],"rates.":[113],"The":[114],"results":[115],"demonstrate":[116],"that":[117],"achieves":[119],"robust":[120],"performance":[121],"across":[122],"diverse":[123],"scenarios":[126],"while":[127],"requiring":[128],"only":[129],"small":[131],"number":[132],"of":[133,154],"trainable":[134],"parameters.":[135],"Extensive":[136],"analyses,":[137],"including":[138],"ablation,":[139],"visualization,":[140],"sensitivity,":[141],"parameter":[142],"efficiency,":[143],"complete-to-missing":[145],"evaluations,":[146],"further":[147],"verify":[148],"the":[149,155],"effectiveness":[150],"generalization":[152],"capability":[153],"proposed":[156],"framework.":[157]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-09-02T00:00:00"}
