{"id":"https://openalex.org/W7164911369","doi":"https://doi.org/10.48550/arxiv.2606.16601","title":"DifferAD-R1: A Difference-Guided IndustrialAnomaly Localization with Multimodal LargeLanguage Models","display_name":"DifferAD-R1: A Difference-Guided IndustrialAnomaly Localization with Multimodal LargeLanguage Models","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164911369","doi":"https://doi.org/10.48550/arxiv.2606.16601"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.16601","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.16601","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.2606.16601","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5023015242","display_name":"Dingrong Wang","orcid":"https://orcid.org/0009-0005-2407-2337"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Dingrong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138695806","display_name":"Xian Tao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tao, Xian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138756770","display_name":"Zhen Qu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qu, Zhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138723221","display_name":"Hengliang Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Hengliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138730937","display_name":"Xinyi Gong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gong, Xinyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138751257","display_name":"Fei Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Fei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138743041","display_name":"Zhengtao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhengtao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138691177","display_name":"Guiguang Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Guiguang","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":0.7559000253677368,"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":0.7559000253677368,"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":0.04690000042319298,"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/T10220","display_name":"Machine Fault Diagnosis Techniques","score":0.04610000178217888,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.6348999738693237},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5931000113487244},{"id":"https://openalex.org/keywords/anomaly-detection","display_name":"Anomaly detection","score":0.5565000176429749},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.5188000202178955},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5141000151634216},{"id":"https://openalex.org/keywords/core","display_name":"Core (optical fiber)","score":0.42489999532699585},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.3488999903202057}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7200999855995178},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6455000042915344},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.6348999738693237},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5931000113487244},{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.5565000176429749},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5188000202178955},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5141000151634216},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47189998626708984},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.42489999532699585},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3488999903202057},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3312999904155731},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.3230000138282776},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28189998865127563},{"id":"https://openalex.org/C12997251","wikidata":"https://www.wikidata.org/wiki/Q567560","display_name":"Anomaly (physics)","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.26989999413490295}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.16601","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.16601","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.2606.16601","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.16601","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":{"Industrial":[0],"anomaly":[1,91],"localization":[2,101],"aims":[3],"to":[4,69,109,170],"accurately":[5],"identify":[6],"and":[7,133,166],"localize":[8],"abnormal":[9],"regions":[10],"in":[11,23,143],"industrial":[12,90,145],"products,":[13],"addressing":[14],"the":[15,51,100],"critical":[16],"challenge":[17],"of":[18,54],"detecting":[19],"unseen":[20],"defect":[21],"categories":[22],"real-world":[24,144],"scenarios.":[25],"Traditional":[26],"closed-set":[27],"methods":[28],"often":[29],"suffer":[30],"from":[31],"poor":[32],"cross-scenario":[33],"generalization,":[34],"while":[35],"existingMultimodal":[36],"Large":[37],"Language":[38],"Model":[39],"(MLLM)-based":[40],"approachesface":[41],"two":[42],"core":[43],"limitations:":[44],"they":[45],"either":[46],"adopt":[47],"QA-style":[48],"paradigmsmisaligned":[49],"with":[50,130],"practical":[52],"demands":[53],"localization,":[55],"or":[56],"relyon":[57],"standard":[58],"optimization":[59,122],"techniques":[60],"such":[61],"as":[62,103],"Group":[63],"RelativePolicy":[64],"Optimization":[65],"(GRPO),":[66],"which":[67],"fails":[68],"deliver":[70],"effectivelearning":[71],"signals":[72],"for":[73,89],"subtle":[74],"defects.":[75],"To":[76,141],"tackle":[77],"these":[78],"issues,":[79],"thispaper":[80],"proposes":[81],"DifferAD-R1,":[82],"an":[83],"MLLM-augmented":[84],"reinforcement":[85],"learning":[86,137],"framework":[87],"tailored":[88],"localization.":[92],"We":[93],"design":[94],"a":[95,104,128],"Difference-Guided":[96],"dual-image":[97],"paradigm,which":[98],"reformulates":[99],"task":[102],"one-shot":[105],"difference":[106],"grounding":[107],"problem":[108],"effectively":[110],"explore":[111],"cross-scenarioanomalies.":[112],"A":[113],"Dual-Consistency":[114],"Localization":[115],"Reward":[116],"is":[117,178],"developedfor":[118],"hard-to-detect":[119],"anomalies,":[120],"enhancing":[121],"stabilityand":[123],"robustness.":[124],"Additionally,":[125],"we":[126,147],"integrate":[127],"difficulty-awarestrategy":[129],"adaptive":[131],"reweighting":[132],"group-wise":[134],"resamplingto":[135],"prioritize":[136],"on":[138],"challenging":[139],"instances.":[140],"facilitateevaluations":[142],"settings,":[146],"construct":[148],"theAD-DualDiff":[149],"dataset,":[150],"comprising":[151],"13K":[152],"paired":[153],"images":[154],"across":[155],"20categories.":[156],"Experimental":[157],"results":[158],"demonstrate":[159],"that":[160],"DifferADR1":[161],"significantly":[162],"outperforms":[163],"existing":[164],"baselines":[165],"achievescompetitive":[167],"performance":[168],"compared":[169],"large-scale":[171],"models":[172],"likeQwen3-VL":[173],"(235B":[174],"parameters).":[175],"Our":[176],"code":[177],"publicly":[179],"availableat:":[180],"https://github.com/Rong2026/work-1.":[181]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-17T00:00:00"}