{"id":"https://openalex.org/W7204533374","doi":"https://doi.org/10.48550/arxiv.2608.26958","title":"Scaling Model-Generated Distillation Data Can Make Latent Teacher Traits More Recoverable","display_name":"Scaling Model-Generated Distillation Data Can Make Latent Teacher Traits More Recoverable","publication_year":2026,"publication_date":"2026-08-27","ids":{"openalex":"https://openalex.org/W7204533374","doi":"https://doi.org/10.48550/arxiv.2608.26958"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.26958","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26958","pdf_url":"https://arxiv.org/pdf/2608.26958","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.26958","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148475563","display_name":"Zhichen Dong","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dong, Zhichen","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory","Shanghai Jiao Tong University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]},{"raw_affiliation_string":"Shanghai Jiao Tong University ,","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148533924","display_name":"Zhixuan Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Zhixuan","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory","Shanghai Jiao Tong University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]},{"raw_affiliation_string":"Shanghai Jiao Tong University ,","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148474934","display_name":"Yuyu Fan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]},{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fan, Yuyu","raw_affiliation_strings":["Fudan University ,","Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University ,","institution_ids":["https://openalex.org/I24943067"]},{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148543610","display_name":"Xiangtian Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li, Xiangtian","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147949240","display_name":"Shuyang Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Shuyang","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5120796872","display_name":"Chao Yang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yang, Chao","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5644999742507935,"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":0.5644999742507935,"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":0.0989999994635582,"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/T10028","display_name":"Topic Modeling","score":0.04520000144839287,"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/knowledge-distillation","display_name":"knowledge distillation","score":0.8619999885559082},{"id":"https://openalex.org/keywords/subliminal-learning","display_name":"subliminal learning","score":0.7940000295639038},{"id":"https://openalex.org/keywords/data-scaling","display_name":"data scaling","score":0.5070000290870667},{"id":"https://openalex.org/keywords/cross-model-transferability","display_name":"cross-model transfer","score":0.6119999885559082},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.414000004529953}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.7883999943733215},{"id":"https://openalex.org/C106934330","wikidata":"https://www.wikidata.org/wiki/Q1971873","display_name":"Trait","level":2,"score":0.7060999870300293},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6559000015258789},{"id":"https://openalex.org/C2780719617","wikidata":"https://www.wikidata.org/wiki/Q1030752","display_name":"Salient","level":2,"score":0.5666000247001648},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4674000144004822},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4115999937057495},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3750999867916107},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.31060001254081726},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.2754000127315521},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2720000147819519},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.26080000400543213}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.26958","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26958","pdf_url":"https://arxiv.org/pdf/2608.26958","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.26958","doi":"https://doi.org/10.48550/arxiv.2608.26958","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.26958","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:2608.26958","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26958","pdf_url":"https://arxiv.org/pdf/2608.26958","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.8278999924659729,"display_name":"Quality education"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204533374.pdf","grobid_xml":"https://content.openalex.org/works/W7204533374.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7154461809","https://openalex.org/W7206186785","https://openalex.org/W7164327085","https://openalex.org/W4417434172","https://openalex.org/W7223271060","https://openalex.org/W7214772452","https://openalex.org/W7221510522","https://openalex.org/W7223321910","https://openalex.org/W7220735973","https://openalex.org/W7215086577"],"abstract_inverted_index":{"Scaling":[0],"model-generated":[1],"data":[2,81,153,190,202],"is":[3,98],"usually":[4,128],"viewed":[5],"as":[6,70],"improving":[7],"distillation:":[8],"more":[9,110,152],"examples":[10,41],"should":[11,191],"increase":[12],"coverage,":[13],"reduce":[14],"noise,":[15],"and":[16,44,180,197],"produce":[17],"stronger":[18],"students.":[19],"We":[20],"show":[21,166],"a":[22,50,57,62,85,147,167],"second":[23],"effect:":[24],"larger":[25,100],"datasets":[26,102],"can":[27,154],"make":[28,103],"subtle":[29],"teacher-specific":[30],"signals":[31],"easier":[32],"to":[33,60],"detect":[34],"in":[35,84,112],"the":[36,47,104,113,126,132,137,158,201],"trained":[37,74],"student,":[38],"even":[39,199],"when":[40,144,200],"are":[42,82],"off-task":[43,67,80,204],"never":[45],"mention":[46],"trait.":[48,160],"In":[49],"controlled":[51],"setup":[52],"inspired":[53],"by":[54],"subliminal":[55],"learning,":[56],"teacher":[58],"induced":[59,106],"express":[61],"target":[63,127],"trait":[64,107,176],"generates":[65],"restricted":[66],"data,":[68],"such":[69],"number-only":[71],"completions.":[72],"Students":[73],"on":[75],"different":[76],"amounts":[77],"of":[78,162],"independent":[79,101],"evaluated":[83],"separate":[86],"domain,":[87],"with":[88,123,194],"matched":[89],"no-trait":[90],"controls":[91],"isolating":[92],"target-specific":[93],"transfer.":[94,182],"Our":[95,183],"main":[96],"finding":[97],"that":[99,142,186],"teacher's":[105],"stand":[108],"out":[109],"clearly":[111],"student's":[114],"later":[115],"behavior.":[116],"Other":[117],"plausible":[118],"traits":[119],"may":[120],"also":[121],"strengthen":[122],"scale,":[124],"but":[125],"grows":[129],"more.":[130],"When":[131],"small-scale":[133],"student":[134],"already":[135],"favors":[136,146],"target,":[138],"scaling":[139,187],"mainly":[140],"amplifies":[141],"behavior;":[143],"it":[145],"related":[148],"or":[149,205],"salient":[150],"alternative,":[151],"shift":[155],"behavior":[156],"toward":[157],"intended":[159],"Analyses":[161],"learned":[163],"LoRA":[164],"updates":[165],"parallel":[168],"trend.":[169],"These":[170],"effects":[171],"appear":[172],"across":[173],"model":[174],"families,":[175],"types,":[177],"multi-trait":[178],"settings,":[179],"cross-model":[181],"results":[184],"suggest":[185],"generated":[188],"distillation":[189],"be":[192],"paired":[193],"trait-aware":[195],"curation":[196],"evaluation,":[198],"appears":[203],"benign.":[206]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-29T00:00:00"}