{"id":"https://openalex.org/W7114805873","doi":"https://doi.org/10.48550/arxiv.2512.08606","title":"Decoupling Template Bias in CLIP: Harnessing Empty Prompts for Enhanced Few-Shot Learning","display_name":"Decoupling Template Bias in CLIP: Harnessing Empty Prompts for Enhanced Few-Shot Learning","publication_year":2025,"publication_date":"2025-12-09","ids":{"openalex":"https://openalex.org/W7114805873","doi":"https://doi.org/10.48550/arxiv.2512.08606"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2512.08606","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.08606","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.2512.08606","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Zhang, Zhenyu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhenyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Chen, Guangyao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Guangyao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zou, Yixiong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zou, Yixiong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Huang, Zhimeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Zhimeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Li, Yuhua","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yuhua","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":0.9280999898910522,"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":0.9280999898910522,"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":0.023399999365210533,"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/T11775","display_name":"COVID-19 diagnosis using AI","score":0.009999999776482582,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6899999976158142},{"id":"https://openalex.org/keywords/offset","display_name":"Offset (computer science)","score":0.6872000098228455},{"id":"https://openalex.org/keywords/decoupling","display_name":"Decoupling (probability)","score":0.4569000005722046},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.36010000109672546},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.34200000762939453},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.3061000108718872}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7364000082015991},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6899999976158142},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.6872000098228455},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6104000210762024},{"id":"https://openalex.org/C205606062","wikidata":"https://www.wikidata.org/wiki/Q5249645","display_name":"Decoupling (probability)","level":2,"score":0.4569000005722046},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.414900004863739},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.36010000109672546},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.34200000762939453},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3077000081539154},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3061000108718872},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.27489998936653137},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.26080000400543213},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2574000060558319},{"id":"https://openalex.org/C83248878","wikidata":"https://www.wikidata.org/wiki/Q344000","display_name":"Active appearance model","level":3,"score":0.2547999918460846}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2512.08606","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.08606","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.2512.08606","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.08606","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":[{"id":"https://metadata.un.org/sdg/4","score":0.5274993777275085,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,88,153],"Contrastive":[1],"Language-Image":[2],"Pre-Training":[3],"(CLIP)":[4],"model":[5,41,123],"excels":[6],"in":[7,57],"few-shot":[8,107],"learning":[9],"by":[10,144],"aligning":[11],"visual":[12,127],"and":[13,31,55,84,98,118,150],"textual":[14,67],"representations.":[15],"Our":[16],"study":[17],"shows":[18],"that":[19,63,69,134],"template-sample":[20],"similarity":[21],"(TSS),":[22],"defined":[23],"as":[24],"the":[25,40,71,103,122],"resemblance":[26],"between":[27,116],"a":[28,61,109],"text":[29],"template":[30,45,82,136],"an":[32],"image":[33],"sample,":[34],"introduces":[35],"bias.":[36,87],"This":[37],"bias":[38,101,110],"leads":[39],"to":[42],"rely":[43],"on":[44,125],"proximity":[46],"rather":[47],"than":[48],"true":[49],"sample-to-category":[50],"alignment,":[51],"reducing":[52],"both":[53],"accuracy":[54,149],"robustness":[56],"classification.":[58],"We":[59],"present":[60],"framework":[62,89],"uses":[64],"empty":[65,95],"prompts,":[66],"inputs":[68],"convey":[70],"idea":[72],"of":[73,155],"\"emptiness\"":[74],"without":[75],"category":[76],"information.":[77],"These":[78],"prompts":[79,96],"capture":[80],"unbiased":[81],"features":[83],"offset":[85],"TSS":[86],"employs":[90],"two":[91],"stages.":[92],"During":[93,106],"pre-training,":[94],"reveal":[97],"reduce":[99],"template-induced":[100],"within":[102],"CLIP":[104],"encoder.":[105],"fine-tuning,":[108],"calibration":[111],"loss":[112],"enforces":[113],"correct":[114],"alignment":[115],"images":[117],"their":[119],"categories,":[120],"ensuring":[121],"focuses":[124],"relevant":[126],"cues.":[128],"Experiments":[129],"across":[130],"multiple":[131],"benchmarks":[132],"demonstrate":[133],"our":[135],"correction":[137],"method":[138],"significantly":[139],"reduces":[140],"performance":[141],"fluctuations":[142],"caused":[143],"TSS,":[145],"yielding":[146],"higher":[147],"classification":[148],"stronger":[151],"robustness.":[152],"repository":[154],"this":[156],"project":[157],"is":[158],"available":[159],"at":[160],"https://github.com/zhenyuZ-HUST/Decoupling-Template-Bias-in-CLIP.":[161]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2025-12-11T00:00:00"}