{"id":"https://openalex.org/W7130559500","doi":"https://doi.org/10.48550/arxiv.2602.16689","title":"Are Object-Centric Representations Better At Compositional Generalization?","display_name":"Are Object-Centric Representations Better At Compositional Generalization?","publication_year":2026,"publication_date":"2026-02-18","ids":{"openalex":"https://openalex.org/W7130559500","doi":"https://doi.org/10.48550/arxiv.2602.16689"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.16689","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058626028","display_name":"Ferdinand Kapl","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Kapl, Ferdinand","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018101784","display_name":"Amir Mohammad Karimi Mamaghan","orcid":"https://orcid.org/0000-0002-6820-948X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mamaghan, Amir Mohammad Karimi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039512035","display_name":"Maximilian Seitzer","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Seitzer, Maximilian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045975901","display_name":"Karl Henrik Johansson","orcid":"https://orcid.org/0000-0001-9940-5929"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Johansson, Karl Henrik","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125126449","display_name":"Carsten Marr","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Marr, Carsten","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126444706","display_name":"Stefan Bauer","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bauer, Stefan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5008956185","display_name":"Andrea Dittadi","orcid":"https://orcid.org/0000-0002-0670-9251"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dittadi, Andrea","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5058626028"],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.597599983215332,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.597599983215332,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.09910000115633011,"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":0.09730000048875809,"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/generalization","display_name":"Generalization","score":0.8034999966621399},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.6136999726295471},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.6022999882698059},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5676000118255615},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4677000045776367},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.43790000677108765},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.4327000081539154},{"id":"https://openalex.org/keywords/data-set","display_name":"Data set","score":0.38519999384880066},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.37290000915527344}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.8034999966621399},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.6136999726295471},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6085000038146973},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6025999784469604},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.6022999882698059},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5676000118255615},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4677000045776367},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.43790000677108765},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.4327000081539154},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38830000162124634},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.38519999384880066},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.37290000915527344},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.36910000443458557},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.353300005197525},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.3449000120162964},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3343000113964081},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3264000117778778},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.2924000024795532},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.28999999165534973},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.28630000352859497},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C96199812","wikidata":"https://www.wikidata.org/wiki/Q2145290","display_name":"Mental representation","level":3,"score":0.26579999923706055},{"id":"https://openalex.org/C21080849","wikidata":"https://www.wikidata.org/wiki/Q13611879","display_name":"Data point","level":2,"score":0.25279998779296875},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.25200000405311584},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2508000135421753}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.16689","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.16689","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.16689","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":"article"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.16689","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.7623975276947021}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Compositional":[0],"generalization,":[1,42],"the":[2,120],"ability":[3],"to":[4,14,39,68,80],"reason":[5],"about":[6],"novel":[7],"combinations":[8,82],"of":[9,34,83,197],"familiar":[10],"concepts,":[11],"is":[12,50,206],"fundamental":[13],"human":[15],"cognition":[16],"and":[17,66,75,90,107,112,123,152,159,185],"a":[18,29,32,54,88],"critical":[19],"challenge":[20],"for":[21,96],"machine":[22],"learning.":[23],"Object-centric":[24],"(OC)":[25],"representations,":[26],"which":[27],"encode":[28],"scene":[30],"as":[31,119],"set":[33],"objects,":[35],"are":[36,135,163],"often":[37],"argued":[38],"support":[40],"such":[41],"but":[43],"systematic":[44],"evidence":[45],"in":[46,137],"visually":[47],"rich":[48],"settings":[49,151],"limited.":[51],"We":[52,109],"introduce":[53],"Visual":[55],"Question":[56],"Answering":[57],"benchmark":[58],"across":[59],"three":[60],"controlled":[61],"visual":[62],"worlds":[63],"(CLEVRTex,":[64],"Super-CLEVR,":[65],"MOVi-C)":[67],"measure":[69],"how":[70],"well":[71],"vision":[72,117],"encoders,":[73,118],"with":[74,170,182],"without":[76],"object-centric":[77,188],"biases,":[78],"generalize":[79],"unseen":[81],"object":[84],"properties.":[85],"To":[86],"ensure":[87],"fair":[89],"comprehensive":[91],"comparison,":[92],"we":[93],"carefully":[94],"account":[95],"training":[97,200],"data":[98,184,201],"diversity,":[99,202],"sample":[100,165],"size,":[101,103,199],"representation":[102],"downstream":[104,157,204],"model":[105],"capacity,":[106],"compute.":[108],"use":[110],"DINOv2":[111],"SigLIP2,":[113],"two":[114],"widely":[115],"used":[116],"foundation":[121],"models":[122,162],"their":[124],"OC":[125,133,147,161],"counterparts.":[126],"Our":[127],"key":[128],"findings":[129],"reveal":[130],"that":[131],"(1)":[132],"approaches":[134],"superior":[136],"harder":[138],"compositional":[139,192],"generalization":[140,169,193],"settings;":[141],"(2)":[142],"original":[143],"dense":[144,174],"representations":[145,189],"surpass":[146,179],"only":[148,181],"on":[149],"easier":[150],"typically":[153],"require":[154],"substantially":[155],"more":[156,164],"compute;":[158],"(3)":[160],"efficient,":[166],"achieving":[167],"stronger":[168,191],"fewer":[171],"images,":[172],"whereas":[173],"encoders":[175],"catch":[176],"up":[177],"or":[178,203],"them":[180],"sufficient":[183],"diversity.":[186],"Overall,":[187],"offer":[190],"when":[194],"any":[195],"one":[196],"dataset":[198],"compute":[205],"constrained.":[207]},"counts_by_year":[],"updated_date":"2026-05-04T08:30:34.212998","created_date":"2026-02-20T00:00:00"}