{"id":"https://openalex.org/W7160637032","doi":"https://doi.org/10.48550/arxiv.2605.06274","title":"When Labels Have Structure: Improving Image Classification with Hierarchy-Aware Cross-Entropy","display_name":"When Labels Have Structure: Improving Image Classification with Hierarchy-Aware Cross-Entropy","publication_year":2026,"publication_date":"2026-05-07","ids":{"openalex":"https://openalex.org/W7160637032","doi":"https://doi.org/10.48550/arxiv.2605.06274"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.06274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.06274","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.2605.06274","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5109689433","display_name":"April Chan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chan, April","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135691986","display_name":"Davide D'Ascenzo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"D'Ascenzo, Davide","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135722085","display_name":"Sebastiano Cultrera di Montesano","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"di Montesano, Sebastiano Cultrera","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":0.30720001459121704,"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":0.30720001459121704,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.18529999256134033,"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":0.12049999833106995,"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/hierarchy","display_name":"Hierarchy","score":0.6901999711990356},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.6550999879837036},{"id":"https://openalex.org/keywords/class-hierarchy","display_name":"Class hierarchy","score":0.6060000061988831},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.5547999739646912},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.37779998779296875},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.3564000129699707},{"id":"https://openalex.org/keywords/abstraction","display_name":"Abstraction","score":0.352400004863739}],"concepts":[{"id":"https://openalex.org/C31170391","wikidata":"https://www.wikidata.org/wiki/Q188619","display_name":"Hierarchy","level":2,"score":0.6901999711990356},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6736000180244446},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.6550999879837036},{"id":"https://openalex.org/C2781289151","wikidata":"https://www.wikidata.org/wiki/Q2903989","display_name":"Class hierarchy","level":3,"score":0.6060000061988831},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5942000150680542},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.5547999739646912},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39969998598098755},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.37779998779296875},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.3564000129699707},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.352400004863739},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3269999921321869},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.3246999979019165},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3003000020980835},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.28630000352859497},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.2840999960899353},{"id":"https://openalex.org/C2778067643","wikidata":"https://www.wikidata.org/wiki/Q166507","display_name":"Interval (graph theory)","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.26249998807907104}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.06274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.06274","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.2605.06274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.06274","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":{"Standard":[0],"cross-entropy":[1,38,133],"is":[2],"the":[3,20,47,57,63,72,83,87,90,94,170],"default":[4],"classification":[5],"loss":[6],"across":[7,110],"virtually":[8],"all":[9,16,156,160],"of":[10,74,137,145,167],"machine":[11],"learning,":[12],"yet":[13],"it":[14],"treats":[15],"misclassifications":[17],"equally,":[18],"ignoring":[19],"semantic":[21],"distances":[22],"that":[23,39,68],"a":[24,33,41,142,164],"class":[25,43,64,92],"hierarchy":[26,44,65],"encodes.":[27],"We":[28,96],"propose":[29],"Hierarchy-Aware":[30],"Cross-Entropy":[31],"(HACE),":[32],"drop-in":[34],"replacement":[35],"for":[36],"standard":[37,132],"incorporates":[40],"known":[42],"directly":[45],"into":[46],"loss.":[48],"HACE":[49,98,128,154],"combines":[50],"two":[51,106],"components:":[52],"prediction":[53],"aggregation,":[54],"which":[55,81],"propagates":[56],"model's":[58],"probability":[59],"mass":[60],"upward":[61],"through":[62],"to":[66,93],"ensure":[67],"parent":[69],"nodes":[70],"accumulate":[71],"confidence":[73],"their":[75],"children;":[76],"and":[77,103,115,118],"ancestral":[78],"label":[79],"smoothing,":[80],"distributes":[82],"ground-truth":[84],"signal":[85],"along":[86],"path":[88],"from":[89],"true":[91],"root.":[95],"evaluate":[97],"on":[99,121,150,159],"CIFAR-100,":[100],"FGVC":[101],"Aircraft,":[102],"NABirds":[104],"in":[105,134],"regimes:":[107],"end-to-end":[108,126],"training":[109],"six":[111],"architectures":[112],"spanning":[113],"convolutional":[114],"attention-based":[116],"designs,":[117],"linear":[119,148],"probing":[120,149],"frozen":[122,151],"DINOv2-Large":[123,152],"features.":[124],"In":[125,147],"training,":[127],"improves":[129],"accuracy":[130],"over":[131,169],"15":[135],"out":[136],"18":[138],"architecture--dataset":[139],"pairs,":[140],"with":[141,163],"mean":[143,165],"gain":[144],"4.66\\%.":[146],"features,":[153],"outperforms":[155],"competing":[157],"methods":[158],"three":[161],"datasets,":[162],"improvement":[166],"2.18\\%":[168],"next":[171],"best":[172],"baseline.":[173]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-09T00:00:00"}