{"id":"https://openalex.org/W7130548588","doi":"https://doi.org/10.48550/arxiv.2602.16093","title":"Updating Parametric Knowledge with Context Distillation Retains Post-Training Capabilities","display_name":"Updating Parametric Knowledge with Context Distillation Retains Post-Training Capabilities","publication_year":2026,"publication_date":"2026-02-17","ids":{"openalex":"https://openalex.org/W7130548588","doi":"https://doi.org/10.48550/arxiv.2602.16093"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2602.16093","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.16093","pdf_url":"https://arxiv.org/pdf/2602.16093","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/2602.16093","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017903527","display_name":"Shankar Padmanabhan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Padmanabhan, Shankar","raw_affiliation_strings":["Department of Computer Science , Cornell University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science , Cornell University","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074059774","display_name":"Mustafa Omer Gul","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gul, Mustafa Omer","raw_affiliation_strings":["Department of Computer Science , Cornell University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science , Cornell University","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020127239","display_name":"Tanya Goyal","orcid":"https://orcid.org/0009-0009-6429-278X","observed_orcids":["https://orcid.org/0009-0009-6429-278X"]},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Goyal, Tanya","raw_affiliation_strings":["Department of Computer Science , Cornell University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science , Cornell University","institution_ids":["https://openalex.org/I205783295"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I205783295"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"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":0.8389000296592712,"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":0.8389000296592712,"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":0.1468999981880188,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.002300000051036477,"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-forgetting","display_name":"knowledge forgetting","score":0.6790000200271606},{"id":"https://openalex.org/keywords/continual-learning","display_name":"continual learning","score":0.48399999737739563},{"id":"https://openalex.org/keywords/kl-divergence-minimization","display_name":"KL divergence minimization","score":0.6769999861717224},{"id":"https://openalex.org/keywords/instruction-following","display_name":"instruction following","score":0.7099999785423279},{"id":"https://openalex.org/keywords/reasoning","display_name":"reasoning","score":0.6129999756813049},{"id":"https://openalex.org/keywords/factual-knowledge","display_name":"factual knowledge","score":0.8339999914169312}],"concepts":[{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.8533999919891357},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.675599992275238},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.636900007724762},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.6328999996185303},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6207000017166138},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6133999824523926},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.588699996471405},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5601000189781189},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.5458999872207642},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.4700999855995178},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.45159998536109924},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.3962000012397766},{"id":"https://openalex.org/C2777220311","wikidata":"https://www.wikidata.org/wiki/Q6423340","display_name":"Knowledge acquisition","level":2,"score":0.3808000087738037},{"id":"https://openalex.org/C2776505523","wikidata":"https://www.wikidata.org/wiki/Q4785468","display_name":"Plan (archaeology)","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.2702000141143799},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.25850000977516174}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2602.16093","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.16093","pdf_url":"https://arxiv.org/pdf/2602.16093","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.2602.16093","doi":"https://doi.org/10.48550/arxiv.2602.16093","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.16093","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2602.16093","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.16093","pdf_url":"https://arxiv.org/pdf/2602.16093","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":[{"display_name":"Quality education","id":"https://metadata.un.org/sdg/4","score":0.4156000018119812}],"study_designs":[],"awards":[{"id":"https://openalex.org/G269386608","display_name":"Collaborative Research: RI: Small: Evaluation Concepts for Assessing and Improving Large Language Models","funder_award_id":"2433072","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7130548588.pdf","grobid_xml":"https://content.openalex.org/works/W7130548588.grobid-xml"},"referenced_works_count":28,"referenced_works":["https://openalex.org/W2560647685","https://openalex.org/W2734314755","https://openalex.org/W2899063268","https://openalex.org/W2902456977","https://openalex.org/W3206867351","https://openalex.org/W3215698779","https://openalex.org/W4282980384","https://openalex.org/W4283451457","https://openalex.org/W4300978888","https://openalex.org/W4306808908","https://openalex.org/W4361001766","https://openalex.org/W4388718089","https://openalex.org/W4388926373","https://openalex.org/W4399990959","https://openalex.org/W4401306886","https://openalex.org/W4403853630","https://openalex.org/W4404782881","https://openalex.org/W4405655184","https://openalex.org/W4412944505","https://openalex.org/W4414281281","https://openalex.org/W4414587625","https://openalex.org/W4414789165","https://openalex.org/W4414939355","https://openalex.org/W4415796274","https://openalex.org/W4417129404","https://openalex.org/W7080122776","https://openalex.org/W7133199932","https://openalex.org/W7133236469"],"related_works":["https://openalex.org/W7133199932","https://openalex.org/W7203597011","https://openalex.org/W7166836333","https://openalex.org/W7223223688","https://openalex.org/W7167936726","https://openalex.org/W7133345210","https://openalex.org/W7128814765","https://openalex.org/W7077069873","https://openalex.org/W7169874691","https://openalex.org/W4416245201"],"abstract_inverted_index":{"Post-training":[0],"endows":[1],"pretrained":[2],"LLMs":[3,18],"with":[4],"a":[5,24,61],"variety":[6],"of":[7,47,80,141],"desirable":[8],"skills,":[9],"including":[10],"instruction-following,":[11,146],"reasoning,":[12,147],"and":[13,43,72,84,114,122,138,148],"others.":[14],"However,":[15],"these":[16],"post-trained":[17,112],"only":[19],"encode":[20],"knowledge":[21,37,68,137],"up":[22],"to":[23,96,119],"cut-off":[25],"date,":[26],"necessitating":[27],"continual":[28,67,126],"adaptation.":[29,69],"Unfortunately,":[30],"existing":[31],"solutions":[32],"cannot":[33],"simultaneously":[34],"learn":[35],"new":[36,136],"from":[38],"an":[39],"adaptation":[40,116],"document":[41],"corpora":[42],"mitigate":[44],"the":[45,81,86,90,131],"forgetting":[46,140],"earlier":[48],"learned":[49,143],"capabilities.":[50],"To":[51],"address":[52],"this,":[53],"we":[54],"introduce":[55],"Distillation":[56],"via":[57],"Split":[58],"Contexts":[59],"(DiSC),":[60],"simple":[62],"context-distillation":[63,99],"based":[64],"approach":[65],"for":[66,125],"\\methodname~derives":[70],"student":[71],"teacher":[73],"distributions":[74],"by":[75],"conditioning":[76],"on":[77,110],"distinct":[78],"segments":[79],"training":[82],"example":[83],"minimizes":[85],"KL":[87],"divergence":[88],"between":[89,134],"shared":[91],"tokens.":[92],"This":[93],"allows":[94],"us":[95],"efficiently":[97],"apply":[98],"without":[100],"requiring":[101],"explicit":[102],"generation":[103],"steps":[104],"during":[105],"training.":[106],"We":[107],"run":[108],"experiments":[109],"four":[111],"models":[113],"two":[115],"domains.":[117],"Compared":[118],"prior":[120],"finetuning":[121],"distillation":[123],"methods":[124],"adaptation,":[127],"DiSC":[128],"consistently":[129],"reports":[130],"best":[132],"trade-off":[133],"learning":[135],"mitigating":[139],"previously":[142],"skills":[144],"like":[145],"factual":[149],"knowledge.":[150]},"counts_by_year":[{"year":2026,"cited_by_count":3}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-02-20T00:00:00"}