{"id":"https://openalex.org/W7160928351","doi":"https://doi.org/10.48550/arxiv.2605.08144","title":"NoiseRater: Meta-Learned Noise Valuation for Diffusion Model Training","display_name":"NoiseRater: Meta-Learned Noise Valuation for Diffusion Model Training","publication_year":2026,"publication_date":"2026-05-02","ids":{"openalex":"https://openalex.org/W7160928351","doi":"https://doi.org/10.48550/arxiv.2605.08144"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.08144","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08144","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.2605.08144","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135970230","display_name":"Fang Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Fang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135990948","display_name":"Haokai Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Haokai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135977861","display_name":"Da Xing","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xing, Da","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036649546","display_name":"Hanqun Cao","orcid":"https://orcid.org/0000-0002-8104-1845"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Hanqun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135989992","display_name":"Tinson Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Tinson","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135939385","display_name":"Yanchao Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yanchao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135926846","display_name":"Xiangru Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Xiangru","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135929700","display_name":"Zehong Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zehong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093364380","display_name":"Aaron Tu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tu, Aaron","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065649249","display_name":"Kuan Pang","orcid":"https://orcid.org/0000-0003-3338-1738"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pang, Kuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100699282","display_name":"Hanchen Wang","orcid":"https://orcid.org/0000-0003-3158-9586"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Hanchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135957793","display_name":"Hongbin Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Hongbin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124795259","display_name":"Zeqi Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Zeqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135955376","display_name":"Yinxi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yinxi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135920595","display_name":"Peng Xia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xia, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135937083","display_name":"Li Erran Li","orcid":"https://orcid.org/0000-0002-9805-3957"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Li Erran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135929053","display_name":"Molei Tao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tao, Molei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135950455","display_name":"Jure Leskovec","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Leskovec, Jure","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135973370","display_name":"Aditya Joshi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Joshi, Aditya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135970457","display_name":"Yejin Choi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Choi, Yejin","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.6601999998092651,"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"}},"topics":[{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.6601999998092651,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.15770000219345093,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.03700000047683716,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.6198999881744385},{"id":"https://openalex.org/keywords/parametric-statistics","display_name":"Parametric statistics","score":0.48829999566078186},{"id":"https://openalex.org/keywords/weighting","display_name":"Weighting","score":0.4839000105857849},{"id":"https://openalex.org/keywords/valuation","display_name":"Valuation (finance)","score":0.4672999978065491},{"id":"https://openalex.org/keywords/noise-measurement","display_name":"Noise measurement","score":0.3458999991416931},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.3407999873161316},{"id":"https://openalex.org/keywords/a-weighting","display_name":"A-weighting","score":0.3379000127315521},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.32749998569488525}],"concepts":[{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.6198999881744385},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6108999848365784},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.48829999566078186},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.4839000105857849},{"id":"https://openalex.org/C186027771","wikidata":"https://www.wikidata.org/wiki/Q4008379","display_name":"Valuation (finance)","level":2,"score":0.4672999978065491},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.3458999991416931},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.3407999873161316},{"id":"https://openalex.org/C70136482","wikidata":"https://www.wikidata.org/wiki/Q13583781","display_name":"A-weighting","level":3,"score":0.3379000127315521},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.32749998569488525},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.3203999996185303},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.32030001282691956},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3197000026702881},{"id":"https://openalex.org/C24574437","wikidata":"https://www.wikidata.org/wiki/Q7135228","display_name":"Parametric model","level":3,"score":0.3183000087738037},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.311599999666214},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3089999854564667},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.29899999499320984},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.26499998569488525},{"id":"https://openalex.org/C100675267","wikidata":"https://www.wikidata.org/wiki/Q1371624","display_name":"Background noise","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.08144","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08144","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.2605.08144","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08144","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Diffusion":[0],"models":[1],"have":[2],"achieved":[3],"remarkable":[4],"success":[5],"across":[6],"a":[7,34,47,94,144],"wide":[8],"range":[9],"of":[10,67],"generative":[11],"tasks,":[12],"yet":[13],"their":[14],"training":[15,69,133],"paradigm":[16],"largely":[17],"treats":[18],"injected":[19],"noise":[20,39,49,57,107,122,130,141],"as":[21,143],"uniformly":[22],"informative.":[23],"In":[24],"this":[25,29],"work,":[26],"we":[27,91],"challenge":[28],"assumption":[30],"and":[31,62,116,126,135,146],"introduce":[32],"NoiseRater,":[33],"meta-learning":[35],"framework":[36],"for":[37,150],"instance-level":[38],"valuation":[40,142],"in":[41],"diffusion":[42,85,152],"model":[43,153],"training.":[44,111,154],"We":[45],"propose":[46],"parametric":[48],"rater":[50,72],"that":[51,98,119,127],"assigns":[52],"importance":[53],"scores":[54],"to":[55,78,105],"individual":[56],"realizations":[58],"conditioned":[59],"on":[60,114],"data":[61],"timestep,":[63],"enabling":[64],"adaptive":[65],"reweighting":[66],"the":[68],"objective.":[70],"The":[71],"is":[73,157],"trained":[74],"via":[75],"bilevel":[76],"optimization":[77],"improve":[79],"downstream":[80],"validation":[81],"performance":[82],"after":[83],"inner-loop":[84],"updates.":[86],"To":[87],"enable":[88],"efficient":[89],"deployment,":[90],"further":[92],"design":[93],"decoupled":[95],"two-stage":[96],"pipeline":[97],"transitions":[99],"from":[100],"soft":[101],"weighting":[102],"during":[103,109],"meta-training":[104],"hard":[106],"selection":[108],"standard":[110],"Extensive":[112],"experiments":[113],"FFHQ":[115],"ImageNet":[117],"demonstrate":[118],"not":[120],"all":[121],"samples":[123],"contribute":[124],"equally,":[125],"prioritizing":[128],"informative":[129],"improves":[131],"both":[132],"efficiency":[134],"generation":[136],"quality.":[137],"Our":[138,155],"results":[139],"establish":[140],"complementary":[145],"previously":[147],"underexplored":[148],"axis":[149],"improving":[151],"code":[156],"available":[158],"at:":[159],"https://anonymous.4open.science/r/NoiseRater-DEB116.":[160]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-13T00:00:00"}