{"id":"https://openalex.org/W7162674804","doi":"https://doi.org/10.48550/arxiv.2605.27469","title":"Architecture-driven Shift: towards a lightweight selector for capturing the trends of logit shift","display_name":"Architecture-driven Shift: towards a lightweight selector for capturing the trends of logit shift","publication_year":2026,"publication_date":"2026-05-26","ids":{"openalex":"https://openalex.org/W7162674804","doi":"https://doi.org/10.48550/arxiv.2605.27469"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.27469","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27469","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.2605.27469","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137277277","display_name":"Zhong Ye","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye, Zhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137227621","display_name":"Yu Hu","orcid":"https://orcid.org/0009-0000-7551-9593"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137291329","display_name":"Ruilin Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Ruilin","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":0.7426000237464905,"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":0.7426000237464905,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.04230000078678131,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.03819999843835831,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/logit","display_name":"Logit","score":0.823199987411499},{"id":"https://openalex.org/keywords/proxy","display_name":"Proxy (statistics)","score":0.5827999711036682},{"id":"https://openalex.org/keywords/logistic-regression","display_name":"Logistic regression","score":0.5824000239372253},{"id":"https://openalex.org/keywords/mixed-logit","display_name":"Mixed logit","score":0.5199000239372253},{"id":"https://openalex.org/keywords/paradigm-shift","display_name":"Paradigm shift","score":0.4551999866962433},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.33169999718666077},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3262999951839447},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.30079999566078186}],"concepts":[{"id":"https://openalex.org/C140331021","wikidata":"https://www.wikidata.org/wiki/Q1868104","display_name":"Logit","level":2,"score":0.823199987411499},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6151999831199646},{"id":"https://openalex.org/C2780148112","wikidata":"https://www.wikidata.org/wiki/Q1432581","display_name":"Proxy (statistics)","level":2,"score":0.5827999711036682},{"id":"https://openalex.org/C151956035","wikidata":"https://www.wikidata.org/wiki/Q1132755","display_name":"Logistic regression","level":2,"score":0.5824000239372253},{"id":"https://openalex.org/C95057490","wikidata":"https://www.wikidata.org/wiki/Q6883984","display_name":"Mixed logit","level":3,"score":0.5199000239372253},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.5020999908447266},{"id":"https://openalex.org/C43540301","wikidata":"https://www.wikidata.org/wiki/Q689971","display_name":"Paradigm shift","level":2,"score":0.4551999866962433},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.33169999718666077},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3314000070095062},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3262999951839447},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.30169999599456787},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.29109999537467957},{"id":"https://openalex.org/C191795146","wikidata":"https://www.wikidata.org/wiki/Q3878446","display_name":"Norm (philosophy)","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C91682802","wikidata":"https://www.wikidata.org/wiki/Q620538","display_name":"Multidimensional scaling","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C61722155","wikidata":"https://www.wikidata.org/wiki/Q6667643","display_name":"Logistic model tree","level":3,"score":0.25769999623298645},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25519999861717224},{"id":"https://openalex.org/C190669063","wikidata":"https://www.wikidata.org/wiki/Q5282043","display_name":"Discrete choice","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.2547999918460846}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.27469","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27469","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.2605.27469","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27469","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":[{"display_name":"Sustainable cities and communities","id":"https://metadata.un.org/sdg/11","score":0.44625431299209595}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Continual":[0],"Learning":[1],"(CL)":[2],"is":[3,172,261],"a":[4,20,35,93,164,175,231,252,262],"practical":[5],"paradigm":[6],"to":[7,23,27,65,132],"utilize":[8],"power":[9],"of":[10,71,74,88,141,196,208,255],"deep":[11],"pre-trained":[12,17],"neural":[13],"networks,":[14],"but":[15],"which":[16,56,79,136,185,260],"model":[18,59,113,166,269],"has":[19,114],"better":[21],"ability":[22],"balance":[24],"``Plasticity-Stability\",":[25],"deserving":[26],"be":[28,100],"chosen?":[29],"The":[30],"logit":[31,42,50,106,124,153,177,242],"shift":[32,43,51,107,125,154,178],"serves":[33],"as":[34,145,251],"natural":[36],"proxy":[37,254],"because":[38,70],"it":[39],"represents":[40],"the":[41,49,72,81,112,120,139,152,182,204,209,214,256],"in":[44,218],"CL":[45,268],"scenarios.":[46,276],"However,":[47],"obtaining":[48],"requires":[52],"huge":[53],"computational":[54],"cost,":[55],"hinders":[57],"large-scale":[58],"selection.":[60],"Existing":[61],"theoretical":[62],"analyses":[63],"fail":[64],"offer":[66],"an":[67],"efficient":[68],"alternative":[69],"assumption":[73],"uniform":[75],"hidden":[76],"layer":[77,201],"widths,":[78],"ignores":[80],"structural":[82],"heterogeneity":[83],"(variable":[84],"width":[85],"and":[86,105,129,212,241],"depth)":[87],"real-world":[89],"architectures.":[90],"This":[91],"raises":[92],"critical":[94],"question:":[95],"what":[96],"theoretically":[97],"relationship":[98],"can":[99,150,249],"identified":[101],"between":[102,239],"heterogeneous":[103],"architecture":[104,127],"on":[108,167,181,188,271],"prior":[109,168],"tasks":[110],"(that":[111],"been":[115],"trained":[116],"on)?":[117],"To":[118],"answer":[119],"question,":[121],"we":[122,245],"decouple":[123],"into":[126],"dependency":[128,131],"data":[130,160],"establish":[133],"our":[134],"framework,":[135],"reveals":[137],"that":[138,149,247],"combination":[140],"two":[142],"dependency,":[143],"defined":[144],"Architecture-driven":[146],"Shift":[147],"(ADS),":[148],"capture":[151],"tendency":[155],"well":[156],"computable":[157],"with":[158,174,200],"few":[159],"samples.":[161],"Specifically,":[162],"for":[163,266],"well-optimized":[165],"tasks,":[169],"higher":[170],"ADS":[171,240,248],"associated":[173],"larger":[176],"after":[179],"training":[180],"current":[183],"task,":[184,211],"derived":[186],"based":[187],"three":[189,272],"mechanistic":[190],"components:":[191],"(1)":[192],"spectral":[193],"norm":[194],"scaling":[195],"weight":[197],"matrix":[198],"gradients":[199],"width,":[202],"(2)":[203],"optimization":[205],"path":[206],"length":[207],"new":[210],"(3)":[213],"asymptotic":[215],"task":[216],"conflict":[217],"wide":[219],"networks.":[220],"Extensive":[221],"empirical":[222],"results":[223],"across":[224,274],"more":[225],"than":[226],"175":[227],"diverse":[228],"architectures":[229],"demonstrate":[230,246],"strong":[232],"monotonic":[233],"correlation":[234],"(the":[235],"weakest":[236],"Spearman's":[237],"$r_s=0.731$)":[238],"shift.":[243],"Practically,":[244],"serve":[250],"lightweight":[253],"expected":[257],"calibration":[258],"error,":[259],"widely":[263],"used":[264],"metric":[265],"reliable":[267],"selection,":[270],"datasets":[273],"six":[275]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-29T00:00:00"}