{"id":"https://openalex.org/W7140706322","doi":"https://doi.org/10.48550/arxiv.2603.24528","title":"Cross-Modal Prototype Alignment and Mixing for Training-Free Few-Shot Classification","display_name":"Cross-Modal Prototype Alignment and Mixing for Training-Free Few-Shot Classification","publication_year":2026,"publication_date":"2026-03-25","ids":{"openalex":"https://openalex.org/W7140706322","doi":"https://doi.org/10.48550/arxiv.2603.24528"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.24528","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.24528","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2603.24528","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130714555","display_name":"Dipam Goswami","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Goswami, Dipam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056168149","display_name":"Simone Magistri","orcid":"https://orcid.org/0000-0002-0520-8463"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Magistri, Simone","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130674979","display_name":"Gido M. van de Ven","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"van de Ven, Gido M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130437939","display_name":"Bart\u0142omiej Twardowski","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Twardowski, Bart\u0142omiej","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064029620","display_name":"Andrew D. Bagdanov","orcid":"https://orcid.org/0000-0001-6408-7043"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bagdanov, Andrew D.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130639210","display_name":"Tinne Tuytelaars","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tuytelaars, Tinne","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130710749","display_name":"Joost van de Weijer","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"van de Weijer, Joost","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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.6416000127792358,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.6416000127792358,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1339000016450882,"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/T10028","display_name":"Topic Modeling","score":0.0348999984562397,"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/embedding","display_name":"Embedding","score":0.6963000297546387},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.6139000058174133},{"id":"https://openalex.org/keywords/subspace-topology","display_name":"Subspace topology","score":0.5871999859809875},{"id":"https://openalex.org/keywords/contextual-image-classification","display_name":"Contextual image classification","score":0.5819000005722046},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.545799970626831},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.4740999937057495},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.35580000281333923},{"id":"https://openalex.org/keywords/image-texture","display_name":"Image texture","score":0.3497999906539917}],"concepts":[{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6963000297546387},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6417999863624573},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.6139000058174133},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5914000272750854},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.5871999859809875},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.5819000005722046},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.545799970626831},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.4740999937057495},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.35580000281333923},{"id":"https://openalex.org/C63099799","wikidata":"https://www.wikidata.org/wiki/Q17147001","display_name":"Image texture","level":4,"score":0.3497999906539917},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.3424000144004822},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3375000059604645},{"id":"https://openalex.org/C27438332","wikidata":"https://www.wikidata.org/wiki/Q2873","display_name":"Principal component analysis","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C138777275","wikidata":"https://www.wikidata.org/wiki/Q6884054","display_name":"Mixing (physics)","level":2,"score":0.3086000084877014},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.29030001163482666},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.289900004863739},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.26170000433921814},{"id":"https://openalex.org/C199579030","wikidata":"https://www.wikidata.org/wiki/Q2851778","display_name":"Automatic image annotation","level":4,"score":0.2540000081062317}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.24528","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.24528","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.24528","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.24528","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-language":[0],"models":[1],"(VLMs)":[2],"like":[3,73],"CLIP":[4],"are":[5,37],"trained":[6],"with":[7,28,143,153],"the":[8,34,48,83,91,106,111,121,125,167,176],"objective":[9],"of":[10,41,50,93,124],"aligning":[11],"text":[12,29,55,127,144],"and":[13,54,60,190],"image":[14,20,31,53,84,107,118,135,139,168],"pairs.":[15],"To":[16],"improve":[17,80,147],"CLIP-based":[18],"few-shot":[19,58,199],"classification,":[21],"recent":[22],"works":[23],"have":[24],"observed":[25],"that,":[26],"along":[27],"embeddings,":[30,145],"embeddings":[32],"from":[33,63,105],"training":[35],"set":[36],"an":[38,191],"important":[39],"source":[40],"information.":[42,98],"In":[43,99],"this":[44,62],"work":[45],"we":[46,115],"investigate":[47],"impact":[49],"directly":[51],"mixing":[52,70],"prototypes":[56,71,79,85,119],"for":[57,150],"classification":[59,81,113,200],"analyze":[61],"a":[64,74,132,185],"bias-variance":[65],"perspective.":[66],"We":[67,164,181],"show":[68,165],"that":[69,166,183],"acts":[72],"shrinkage":[75],"estimator.":[76],"Although":[77],"mixed":[78,142,187],"performance,":[82],"still":[86,171],"add":[87],"some":[88],"noise":[89],"in":[90,157],"form":[92],"instance-specific":[94],"background":[95],"or":[96],"context":[97],"order":[100],"to":[101,110,130],"capture":[102],"only":[103],"information":[104],"space":[108,129],"relevant":[109],"given":[112],"task,":[114],"propose":[116],"projecting":[117],"onto":[120],"principal":[122],"directions":[123],"semantic":[126,134,159],"embedding":[128],"obtain":[131],"text-aligned":[133,138,186],"subspace.":[136],"These":[137],"prototypes,":[140],"when":[141],"further":[146],"classification.":[148],"However,":[149],"downstream":[151],"datasets":[152],"poor":[154],"cross-modal":[155],"alignment":[156,160],"CLIP,":[158],"might":[161],"be":[162,172],"suboptimal.":[163],"subspace":[169],"can":[170],"leveraged":[173],"by":[174],"modeling":[175],"anisotropy":[177],"using":[178],"class":[179],"covariances.":[180],"demonstrate":[182],"combining":[184],"prototype":[188],"classifier":[189,194],"image-specific":[192],"LDA":[193],"outperforms":[195],"existing":[196],"methods":[197],"across":[198],"benchmarks.":[201]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-27T00:00:00"}