{"id":"https://openalex.org/W7165372013","doi":"https://doi.org/10.48550/arxiv.2606.20300","title":"CMDS-AD: Cross-Modal Dual-Stream Decoupling for Few-Shot Anomaly Detection","display_name":"CMDS-AD: Cross-Modal Dual-Stream Decoupling for Few-Shot Anomaly Detection","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165372013","doi":"https://doi.org/10.48550/arxiv.2606.20300"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.20300","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.20300","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.2606.20300","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138980117","display_name":"Junhao Cai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cai, Junhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139022046","display_name":"Deyu Zeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Junyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091125853","display_name":"Junhao Pang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Deyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138975520","display_name":"Junyu Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pang, Junhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125530564","display_name":"Qiwei Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Qiwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138980475","display_name":"Xiaopin Zhong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhong, Xiaopin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139009093","display_name":"Zongze Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Zongze","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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.7466999888420105,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.7466999888420105,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.13449999690055847,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.013500000350177288,"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/anomaly-detection","display_name":"Anomaly detection","score":0.742900013923645},{"id":"https://openalex.org/keywords/rgb-color-model","display_name":"RGB color model","score":0.5665000081062317},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5382000207901001},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4733000099658966},{"id":"https://openalex.org/keywords/anomaly","display_name":"Anomaly (physics)","score":0.43970000743865967},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.41350001096725464},{"id":"https://openalex.org/keywords/limit","display_name":"Limit (mathematics)","score":0.382099986076355},{"id":"https://openalex.org/keywords/extreme-point","display_name":"Extreme point","score":0.36469998955726624},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.3531999886035919}],"concepts":[{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.742900013923645},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6277999877929688},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5766000151634216},{"id":"https://openalex.org/C82990744","wikidata":"https://www.wikidata.org/wiki/Q166194","display_name":"RGB color model","level":2,"score":0.5665000081062317},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5382000207901001},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4733000099658966},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4691999852657318},{"id":"https://openalex.org/C12997251","wikidata":"https://www.wikidata.org/wiki/Q567560","display_name":"Anomaly (physics)","level":2,"score":0.43970000743865967},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.41350001096725464},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.382099986076355},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.38190001249313354},{"id":"https://openalex.org/C39847760","wikidata":"https://www.wikidata.org/wiki/Q1385465","display_name":"Extreme point","level":2,"score":0.36469998955726624},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.3531999886035919},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.3529999852180481},{"id":"https://openalex.org/C147581598","wikidata":"https://www.wikidata.org/wiki/Q729429","display_name":"Extreme value theory","level":2,"score":0.3353999853134155},{"id":"https://openalex.org/C205606062","wikidata":"https://www.wikidata.org/wiki/Q5249645","display_name":"Decoupling (probability)","level":2,"score":0.3111000061035156},{"id":"https://openalex.org/C42747912","wikidata":"https://www.wikidata.org/wiki/Q1048447","display_name":"Multiplicative function","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C87360688","wikidata":"https://www.wikidata.org/wiki/Q740686","display_name":"Synthetic aperture radar","level":2,"score":0.2973000109195709},{"id":"https://openalex.org/C132094186","wikidata":"https://www.wikidata.org/wiki/Q641585","display_name":"Clutter","level":3,"score":0.2930999994277954},{"id":"https://openalex.org/C2987395694","wikidata":"https://www.wikidata.org/wiki/Q557891","display_name":"Feature tracking","level":3,"score":0.28600001335144043},{"id":"https://openalex.org/C67226441","wikidata":"https://www.wikidata.org/wiki/Q1665389","display_name":"Robust statistics","level":3,"score":0.27790001034736633},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C43711488","wikidata":"https://www.wikidata.org/wiki/Q7534783","display_name":"Skew","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25839999318122864}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.20300","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.20300","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.2606.20300","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.20300","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Few-shot":[0],"anomaly":[1,11],"detection":[2,12],"remains":[3],"challenging":[4],"due":[5],"to":[6,22,77,140],"limited":[7],"training":[8],"data.":[9],"Multi-modal":[10],"(MAD)":[13],"offers":[14],"a":[15,63,88,93,102,154,188],"viable":[16],"solution,":[17],"leveraging":[18],"3D":[19,83],"geometric":[20],"cues":[21],"enrich":[23],"2D":[24],"RGB":[25,75,112],"representations":[26,110],"and":[27,53,128,137,174,182],"compensate":[28],"for":[29],"this":[30,97],"scarcity.":[31,81],"However,":[32],"existing":[33],"MAD":[34],"methods":[35],"apply":[36],"spatially":[37],"uniform":[38],"feature":[39],"processing,":[40],"conflating":[41],"stable":[42],"macroscopic":[43],"structures":[44],"with":[45],"high-frequency":[46],"localized":[47],"defect":[48],"signals,":[49],"exacerbating":[50],"cross-modal":[51,151],"misalignment":[52],"inflating":[54],"false-positive":[55],"rates.":[56],"To":[57],"overcome":[58],"this,":[59],"we":[60,86],"present":[61],"CMDS-AD,":[62],"Cross-Modal":[64],"Dual-Stream":[65],"Anomaly":[66],"Detection":[67],"framework.":[68],"A":[69,144],"LoRA-guided":[70],"diffusion":[71,90],"model":[72,91],"generates":[73],"diverse":[74],"samples":[76],"mitigate":[78],"extreme":[79,163],"data":[80],"For":[82],"normal":[84,94,109],"augmentation,":[85],"employ":[87],"pre-trained":[89],"as":[92,101],"estimator.":[95],"Crucially,":[96],"estimator":[98],"inherently":[99],"acts":[100],"non-linear":[103],"low-pass":[104],"filter,":[105],"directly":[106],"extracting":[107],"low-frequency":[108,122,138],"from":[111],"inputs.":[113],"This":[114],"establishes":[115],"an":[116],"auxiliary":[117],"estimated":[118],"stream":[119],"of":[120,171],"purely":[121],"information,":[123],"anchoring":[124],"robust":[125],"structural":[126],"templates":[127],"assisting":[129],"the":[130,162],"uncompressed":[131],"real":[132],"stream,":[133],"containing":[134],"coupled":[135],"high-":[136],"components,":[139],"precisely":[141],"isolate":[142],"micro-defects.":[143],"Coordinate-Aware":[145],"Hierarchical":[146],"Feature":[147],"Mapper":[148],"adaptively":[149],"aligns":[150],"semantics,":[152],"while":[153],"multiplicative":[155],"scoring":[156],"mechanism":[157],"filters":[158],"modality-specific":[159],"noise.":[160],"Under":[161],"1-shot":[164],"setting,":[165],"CMDS-AD":[166],"achieves":[167],"absolute":[168],"performance":[169],"gains":[170],"5.7%":[172],"(I-AUROC)":[173],"2.0%":[175],"(AUPRO)":[176],"on":[177,185],"MVTec":[178],"3D-AD,":[179],"alongside":[180],"7.7%":[181],"5.6%":[183],"improvements":[184],"EyeCandies,":[186],"establishing":[187],"new":[189],"state-of-the-art.":[190],"Code":[191],"is":[192],"available":[193],"at":[194],"https://github.com/Junhaocai27/CMDS-AD":[195]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-20T00:00:00"}