{"id":"https://openalex.org/W7130733437","doi":"https://doi.org/10.48550/arxiv.2602.17419","title":"EAGLE: Expert-Augmented Attention Guidance for Tuning-Free Industrial Anomaly Detection in Multimodal Large Language Models","display_name":"EAGLE: Expert-Augmented Attention Guidance for Tuning-Free Industrial Anomaly Detection in Multimodal Large Language Models","publication_year":2026,"publication_date":"2026-02-19","ids":{"openalex":"https://openalex.org/W7130733437","doi":"https://doi.org/10.48550/arxiv.2602.17419"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2602.17419","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2602.17419","pdf_url":"https://arxiv.org/pdf/2602.17419","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2602.17419","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5126517637","display_name":"Xiaomeng Peng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I138925566","display_name":"Ewha Womans University","ror":"https://ror.org/053fp5c05","country_code":"KR","type":"education","lineage":["https://openalex.org/I138925566"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Peng, Xiaomeng","raw_affiliation_strings":["Ewha Womans University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ewha Womans University","institution_ids":["https://openalex.org/I138925566"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022116737","display_name":"Xilang Huang","orcid":"https://orcid.org/0000-0002-1506-1354","observed_orcids":["https://orcid.org/0000-0002-1506-1354"]},"institutions":[{"id":"https://openalex.org/I138925566","display_name":"Ewha Womans University","ror":"https://ror.org/053fp5c05","country_code":"KR","type":"education","lineage":["https://openalex.org/I138925566"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Huang, Xilang","raw_affiliation_strings":["Ewha Womans University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ewha Womans University","institution_ids":["https://openalex.org/I138925566"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5027580730","display_name":"Seon Han Choi","orcid":"https://orcid.org/0000-0003-3332-7723","observed_orcids":["https://orcid.org/0000-0003-3332-7723"]},"institutions":[{"id":"https://openalex.org/I138925566","display_name":"Ewha Womans University","ror":"https://ror.org/053fp5c05","country_code":"KR","type":"education","lineage":["https://openalex.org/I138925566"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Choi, Seon Han","raw_affiliation_strings":["Ewha Womans University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ewha Womans University","institution_ids":["https://openalex.org/I138925566"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I138925566"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":2,"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/T12111","display_name":"Industrial Vision Systems and Defect Detection","score":0.9154999852180481,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12111","display_name":"Industrial Vision Systems and Defect Detection","score":0.9154999852180481,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.07150000333786011,"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":0.0034000000450760126,"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/industrial-anomaly-detection","display_name":"industrial anomaly detection","score":0.9769999980926514},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.996999979019165},{"id":"https://openalex.org/keywords/attentional-guidance","display_name":"attention guidance","score":0.703000009059906},{"id":"https://openalex.org/keywords/visual-defect-detection","display_name":"visual defect detection","score":0.42100000381469727},{"id":"https://openalex.org/keywords/mvtec-ad","display_name":"MVTec AD","score":0.8199999928474426},{"id":"https://openalex.org/keywords/visa","display_name":"Visa","score":0.7850000262260437},{"id":"https://openalex.org/keywords/frozen-language-models","display_name":"frozen language models","score":0.3799999952316284}],"concepts":[{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.8880000114440918},{"id":"https://openalex.org/C12997251","wikidata":"https://www.wikidata.org/wiki/Q567560","display_name":"Anomaly (physics)","level":2,"score":0.7077000141143799},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6834999918937683},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5619000196456909},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42590001225471497},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4120999872684479},{"id":"https://openalex.org/C2780985876","wikidata":"https://www.wikidata.org/wiki/Q2092297","display_name":"Eagle","level":2,"score":0.3734000027179718},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.3379000127315521},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.31439998745918274},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.30660000443458557},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.3012000024318695},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.2955999970436096},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2937000095844269},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.26109999418258667}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2602.17419","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2602.17419","pdf_url":"https://arxiv.org/pdf/2602.17419","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2602.17419","doi":"https://doi.org/10.48550/arxiv.2602.17419","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.17419","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2602.17419","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2602.17419","pdf_url":"https://arxiv.org/pdf/2602.17419","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7130733437.pdf","grobid_xml":"https://content.openalex.org/works/W7130733437.grobid-xml"},"referenced_works_count":15,"referenced_works":["https://openalex.org/W4388093086","https://openalex.org/W4388093160","https://openalex.org/W4393119454","https://openalex.org/W4394709684","https://openalex.org/W4403565644","https://openalex.org/W4403575216","https://openalex.org/W4407794264","https://openalex.org/W4414850555","https://openalex.org/W4415085973","https://openalex.org/W4416047734","https://openalex.org/W4416053382","https://openalex.org/W4416084375","https://openalex.org/W4416542271","https://openalex.org/W6966553164","https://openalex.org/W7131386314"],"related_works":["https://openalex.org/W7169675727","https://openalex.org/W4416535398","https://openalex.org/W7213700358","https://openalex.org/W7154300031","https://openalex.org/W6979294322","https://openalex.org/W4388093160","https://openalex.org/W4415598949","https://openalex.org/W7221146982","https://openalex.org/W7137941890","https://openalex.org/W7133302756"],"abstract_inverted_index":{"Multimodal":[0],"large":[1],"language":[2],"models":[3],"(MLLMs)":[4],"can":[5],"enrich":[6],"industrial":[7],"anomaly":[8,14,21,59,114,149],"detection":[9,25],"with":[10,43,61,118,157],"semantic":[11,164],"descriptions":[12],"and":[13,80,83,86,110,132,146,153],"reasoning,":[15],"but":[16],"they":[17],"still":[18],"lag":[19],"specialist":[20],"detectors":[22,60],"in":[23,148],"binary":[24],"accuracy.":[26],"Existing":[27],"approaches":[28],"address":[29],"this":[30,128],"gap":[31],"by":[32],"fine-tuning":[33],"MLLMs":[34],"or":[35],"training":[36],"bridging":[37],"modules":[38],"to":[39,144],"align":[40],"expert":[41,58,77,99],"outputs":[42],"MLLM":[44,93,108,137,163],"inputs,":[45],"limiting":[46],"flexibility":[47],"across":[48],"backbones.":[49],"We":[50],"propose":[51],"EAGLE,":[52],"a":[53,73],"tuning-free":[54],"framework":[55],"that":[56,112],"integrates":[57],"frozen":[62],"MLLMs.":[63],"EAGLE":[64,125,134],"consists":[65],"of":[66],"Threshold-Guided":[67],"Prompt":[68],"Selection":[69],"(TGPS),":[70],"which":[71,91],"estimates":[72],"decision":[74],"threshold":[75],"from":[76],"model":[78],"statistics":[79],"selects":[81],"textual":[82],"visual":[84,96],"prompts,":[85],"Confidence-Aware":[87],"Attention":[88],"Sharpening":[89],"(CAAS),":[90],"shifts":[92],"attention":[94,109],"toward":[95],"evidence":[97],"when":[98],"confidence":[100],"is":[101],"low.":[102],"Beyond":[103],"improving":[104],"accuracy,":[105,151],"we":[106],"analyze":[107],"find":[111],"correct":[113],"predictions":[115],"are":[116],"associated":[117],"stronger":[119],"focus":[120],"on":[121],"ground-truth":[122],"defect":[123],"regions;":[124],"consistently":[126],"strengthens":[127],"alignment.":[129],"On":[130],"MVTec-AD":[131],"VisA,":[133],"improves":[135],"five":[136],"backbones":[138],"without":[139],"parameter":[140],"updates,":[141],"reaching":[142],"up":[143],"94.4\\%":[145],"88.1\\%":[147],"discrimination":[150],"respectively,":[152],"achieving":[154],"performance":[155],"competitive":[156],"fine-tuning-based":[158],"methods":[159],"while":[160],"largely":[161],"preserving":[162],"reasoning":[165],"ability.":[166]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-02-21T00:00:00"}