{"id":"https://openalex.org/W7165926436","doi":"https://doi.org/10.48550/arxiv.2606.25165","title":"The Gentle Collapse: Distributional Metrics for Continual Learning","display_name":"The Gentle Collapse: Distributional Metrics for Continual Learning","publication_year":2026,"publication_date":"2026-06-23","ids":{"openalex":"https://openalex.org/W7165926436","doi":"https://doi.org/10.48550/arxiv.2606.25165"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.25165","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.25165","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,"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.2606.25165","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136791785","display_name":"Ahmed Anwar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anwar, Ahmed","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007820528","display_name":"Andreas Wagner","orcid":"https://orcid.org/0000-0001-5772-1261"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wagner, Andreas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029509133","display_name":"Federico Raue","orcid":"https://orcid.org/0000-0002-8604-6207"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Raue, Federico","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099038777","display_name":"Tobias Nauen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nauen, Tobias","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139307650","display_name":"Andreas Dengel","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dengel, Andreas","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":0.632099986076355,"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":0.632099986076355,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.05869999900460243,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.052299998700618744,"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/forgetting","display_name":"Forgetting","score":0.8449000120162964},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.7662000060081482},{"id":"https://openalex.org/keywords/margin","display_name":"Margin (machine learning)","score":0.7099999785423279},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.5401999950408936},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.4884999990463257},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.4318000078201294},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.41350001096725464},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.38769999146461487}],"concepts":[{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.8449000120162964},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.7662000060081482},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.7099999785423279},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5512999892234802},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.5401999950408936},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4884999990463257},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4318000078201294},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43140000104904175},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.41350001096725464},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.38769999146461487},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3684000074863434},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3472000062465668},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.34310001134872437},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.32749998569488525},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.3260999917984009},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.32580000162124634},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.31200000643730164},{"id":"https://openalex.org/C2780898871","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance metric","level":2,"score":0.30219998955726624},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C2781140086","wikidata":"https://www.wikidata.org/wiki/Q557945","display_name":"Confusion","level":2,"score":0.27970001101493835},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.26910001039505005},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C21080849","wikidata":"https://www.wikidata.org/wiki/Q13611879","display_name":"Data point","level":2,"score":0.2630000114440918},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.25220000743865967},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.25165","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.25165","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,"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.2606.25165","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.25165","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,"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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Accuracy":[0],"degradation":[1],"is":[2,36,101,165],"the":[3,22,31,80,128,188],"standard":[4],"metric":[5,108,132],"for":[6],"Catastrophic":[7],"Forgetting":[8],"(CF),":[9],"however,":[10],"it":[11],"records":[12],"only":[13],"whether":[14],"forgetting":[15,55,115,182],"occurred":[16],"or":[17],"not.":[18],"It":[19],"saturates":[20],"at":[21,27,77,142],"extremes":[23],"and":[24,50],"collapses":[25],"discretely":[26],"task":[28],"boundaries,":[29],"hiding":[30],"internal":[32],"structure":[33],"of":[34,86,130],"what":[35],"being":[37],"forgotten.":[38],"We":[39,95],"introduce":[40],"six":[41],"softmax-derived":[42],"metrics":[43,70],"spanning":[44],"true-label":[45],"rank":[46],"(TLR),":[47],"predictive":[48],"confidence,":[49],"distributional":[51],"divergence":[52],"that":[53,91,97],"characterize":[54],"continuously,":[56],"each":[57],"normalized":[58],"to":[59,65,151],"[0,":[60],"1]":[61],"with":[62],"no":[63],"modification":[64],"training.":[66],"On":[67,176],"CIFAR-100,":[68],"these":[69],"carry":[71],"information":[72],"where":[73],"accuracy":[74,92],"does":[75],"not:":[76],"0%":[78],"accuracy,":[79],"Confusion":[81],"Margin":[82],"spans":[83],"an":[84],"IQR":[85],"[0.32,":[87],"0.50]":[88],"across":[89],"classes":[90],"treats":[93],"identically.":[94],"demonstrate":[96],"this":[98],"richer":[99],"signal":[100],"actionable":[102],"in":[103],"mitigating":[104],"catastrophic":[105],"forgetting.":[106],"Per-sample":[107],"scores":[109],"used":[110],"as":[111],"loss":[112],"weights":[113],"reduce":[114],"by":[116,183],"1.3":[117],"percentage":[118,185],"points":[119,186],"over":[120,133,187],"uniform":[121],"experience":[122],"replay":[123],"(ER)":[124],"on":[125],"CIFAR-100.":[126],"Furthermore,":[127],"slope":[129],"a":[131,134,138,143,173],"small":[135],"window":[136],"provides":[137],"stable":[139],"sampling":[140,180],"criterion:":[141],"small-window":[144,169],"size":[145],"(e.g.":[146],"3":[147],"epochs),":[148],"accuracy-trend":[149],"degrades":[150],"34.79%":[152],"(std.":[153,160],"=":[154,161],"2.32)":[155],"while":[156],"log-TLR":[157,178],"achieves":[158],"41.07%":[159],"0.57).":[162],"This":[163],"gap":[164],"structural":[166],"since":[167],"reliable":[168],"trend":[170,179],"estimation":[171],"requires":[172],"continuous":[174],"signal.":[175],"TinyImageNet,":[177],"reduces":[181],"7.7":[184],"ER":[189],"baseline.":[190]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-26T00:00:00"}