{"id":"https://openalex.org/W7224084273","doi":"https://doi.org/10.48550/arxiv.2610.36679","title":"Specialization, Diversity, and Scaling in Child-Scale Language Models","display_name":"Specialization, Diversity, and Scaling in Child-Scale Language Models","publication_year":2026,"publication_date":"2026-10-27","ids":{"openalex":"https://openalex.org/W7224084273","doi":"https://doi.org/10.48550/arxiv.2610.36679"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.36679","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.36679","pdf_url":"https://arxiv.org/pdf/2610.36679","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.36679","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039277252","display_name":"Steven Y. Feng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Feng, Steven Y.","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5132713143","display_name":"Michael C. Frank","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Frank, Michael C.","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I97018004"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10730","display_name":"Language Development and Disorders","score":0.9829000234603882,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10730","display_name":"Language Development and Disorders","score":0.9829000234603882,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.005200000014156103,"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/T12031","display_name":"Speech and dialogue systems","score":0.003599999938160181,"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/child-directed-language","display_name":"child-directed language","score":0.6460000276565552},{"id":"https://openalex.org/keywords/language-model-training-data","display_name":"language model training data","score":0.5149999856948853},{"id":"https://openalex.org/keywords/grammatical-generalization","display_name":"grammatical generalization","score":0.6140000224113464},{"id":"https://openalex.org/keywords/synthetic-dialogue-data","display_name":"synthetic dialogue data","score":0.5189999938011169},{"id":"https://openalex.org/keywords/dataset-diversity","display_name":"data diversity","score":0.41100001335144043},{"id":"https://openalex.org/keywords/model-scaling","display_name":"model scaling","score":0.5149999856948853},{"id":"https://openalex.org/keywords/n-gram-language-models","display_name":"n-gram language models","score":0.6010000109672546}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6855000257492065},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.666700005531311},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5730000138282776},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5692999958992004},{"id":"https://openalex.org/C74672266","wikidata":"https://www.wikidata.org/wiki/Q815859","display_name":"Language acquisition","level":2,"score":0.5254999995231628},{"id":"https://openalex.org/C2781202465","wikidata":"https://www.wikidata.org/wiki/Q18346297","display_name":"Lexical diversity","level":3,"score":0.41659998893737793},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.39800000190734863},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3977999985218048},{"id":"https://openalex.org/C39608478","wikidata":"https://www.wikidata.org/wiki/Q5015979","display_name":"Cache language model","level":5,"score":0.3815000057220459},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3783000111579895},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3720000088214874},{"id":"https://openalex.org/C94922259","wikidata":"https://www.wikidata.org/wiki/Q33215","display_name":"Constructed language","level":2,"score":0.357699990272522},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3571999967098236},{"id":"https://openalex.org/C26022165","wikidata":"https://www.wikidata.org/wiki/Q8091","display_name":"Grammar","level":2,"score":0.3481999933719635},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3179999887943268},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.3059000074863434},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2906999886035919},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.28760001063346863},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.2565000057220459},{"id":"https://openalex.org/C2779313563","wikidata":"https://www.wikidata.org/wiki/Q17072565","display_name":"On Language","level":2,"score":0.2533999979496002}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.36679","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.36679","pdf_url":"https://arxiv.org/pdf/2610.36679","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.36679","doi":"https://doi.org/10.48550/arxiv.2610.36679","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.36679","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.36679","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.36679","pdf_url":"https://arxiv.org/pdf/2610.36679","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"score":0.5385000109672546,"display_name":"Quality education","id":"https://metadata.un.org/sdg/4"}],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320309480","display_name":"Nvidia","ror":"https://ror.org/03jdj4y14"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224084273.pdf","grobid_xml":"https://content.openalex.org/works/W7224084273.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7147416778","https://openalex.org/W4404782836","https://openalex.org/W7062738394","https://openalex.org/W4416113801","https://openalex.org/W4404391336","https://openalex.org/W4404782547","https://openalex.org/W7019726408","https://openalex.org/W7137059568","https://openalex.org/W7203780040","https://openalex.org/W4416037635"],"abstract_inverted_index":{"Modern":[0],"language":[1,8,59,99,143,151],"models":[2,27],"are":[3],"trained":[4,55],"on":[5,28,40,56,88,103,113,148,156],"far":[6],"more":[7],"than":[9,66],"children":[10,33,92],"experience,":[11],"but":[12,155],"we":[13],"know":[14],"much":[15,150],"less":[16],"about":[17],"learning":[18,144],"from":[19,34,90,161],"small,":[20],"naturalistic":[21,119],"datasets.":[22],"We":[23],"train":[24],"small":[25],"GPT-2":[26],"transcripts":[29],"of":[30,51,71,96],"speech":[31],"with":[32,131],"dense":[35],"home":[36],"recordings":[37],"(BabyView)":[38],"and":[39,49,76,120,128,162],"synthetic":[41,121],"child-directed":[42],"dialogue":[43],"(TinyDialogues),":[44],"varying":[45],"the":[46,83,169],"amount,":[47],"source,":[48],"diversity":[50],"training":[52,102,122],"data.":[53],"Models":[54],"one":[57],"child's":[58,106],"environment":[60,63],"predict":[61],"that":[62,141,165],"substantially":[64],"better":[65],"other":[67],"children's":[68],"environments.":[69],"Much":[70],"this":[72],"specialization":[73],"is":[74,171],"lexical":[75],"distributional:":[77],"a":[78,104,109,152],"simple":[79],"5-gram":[80],"model":[81,153,170],"reproduces":[82],"cross-child":[84],"pattern":[85],"(r=.91).":[86],"Training":[87],"data":[89,117],"multiple":[91],"clearly":[93],"improves":[94],"prediction":[95],"pooled":[97],"held-out":[98],"relative":[100],"to":[101,173],"single":[105],"input":[107,166],"without":[108],"similarly":[110],"clear":[111],"improvement":[112],"grammatical":[114,127,135],"benchmarks.":[115],"Across":[116],"sizes,":[118],"inputs":[123],"also":[124],"produce":[125],"different":[126],"lexical-semantic":[129],"patterns,":[130],"important":[132],"differences":[133],"between":[134],"phenomena.":[136],"Together,":[137],"these":[138],"results":[139],"show":[140],"small-data":[142],"depends":[145],"not":[146],"only":[147],"how":[149,163],"sees,":[154],"which":[157],"environments":[158],"it":[159],"comes":[160],"well":[164],"matches":[167],"what":[168],"asked":[172],"predict.":[174]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-10-29T00:00:00"}