{"id":"https://openalex.org/W7139962992","doi":"https://doi.org/10.48550/arxiv.2603.18073","title":"Continually self-improving AI","display_name":"Continually self-improving AI","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7139962992","doi":"https://doi.org/10.48550/arxiv.2603.18073"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.18073","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.18073","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":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.18073","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130241630","display_name":"Zitong Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yang, Zitong","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5130241630"],"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/T10181","display_name":"Natural Language Processing Techniques","score":0.11330000311136246,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.11330000311136246,"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":0.07850000262260437,"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/T14347","display_name":"Big Data and Digital Economy","score":0.0737999975681305,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/key","display_name":"Key (lock)","score":0.6263999938964844},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.49140000343322754},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4129999876022339},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.367900013923645},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.3492000102996826},{"id":"https://openalex.org/keywords/test","display_name":"Test (biology)","score":0.33169999718666077},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.3312000036239624}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7490000128746033},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.6263999938964844},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5860000252723694},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.49140000343322754},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41679999232292175},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4129999876022339},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.367900013923645},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.3492000102996826},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.33169999718666077},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.3312000036239624},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.32100000977516174},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.29429998993873596},{"id":"https://openalex.org/C16910744","wikidata":"https://www.wikidata.org/wiki/Q7705759","display_name":"Test data","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2856000065803528},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C2993724205","wikidata":"https://www.wikidata.org/wiki/Q315","display_name":"Human language","level":2,"score":0.2540000081062317}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.18073","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.18073","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":"doi:10.48550/arxiv.2603.18073","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.18073","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":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.7511872053146362}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Modern":[0],"language":[1],"model-based":[2],"AI":[3,65,196],"systems":[4,49],"are":[5,67],"remarkably":[6],"powerful,":[7],"yet":[8],"their":[9,15],"capabilities":[10,168],"remain":[11],"fundamentally":[12],"capped":[13],"by":[14,69,185],"human":[16,73,145,208],"creators":[17],"in":[18,109],"three":[19,91],"key":[20],"ways.":[21],"First,":[22,103],"although":[23],"a":[24,82,114,129,151,200],"model's":[25],"weights":[26],"can":[27,75,159,197,210],"be":[28],"updated":[29],"via":[30],"fine-tuning,":[31],"acquiring":[32],"new":[33],"knowledge":[34,110,126],"from":[35,56,136,171],"small,":[36],"specialized":[37],"corpora":[38,123],"after":[39],"pretraining":[40,167],"remains":[41],"highly":[42],"data-inefficient.":[43],"Second,":[44,140],"the":[45,60,70,157,192],"training":[46,180],"of":[47,154,194,203],"these":[48,87,96],"relies":[50],"heavily":[51],"on":[52,144],"finite,":[53],"human-generated":[54],"data":[55,116,162],"across":[57],"history.":[58],"Third,":[59],"pipelines":[61],"used":[62],"to":[63,98,104,131,141,163,177],"train":[64],"models":[66],"confined":[68],"algorithms":[71],"that":[72,118,149,184],"researchers":[74,209],"discover":[76],"and":[77,120],"explore.":[78],"This":[79],"thesis":[80],"takes":[81],"small":[83,122],"step":[84],"toward":[85],"overcoming":[86],"inherent":[88],"limitations,":[89],"presenting":[90],"chapters":[92],"aimed":[93],"at":[94],"breaking":[95],"dependencies":[97],"create":[99],"continually":[100],"self-improving":[101],"AI.":[102],"overcome":[105],"this":[106],"data-efficiency":[107],"barrier":[108],"acquisition,":[111],"we":[112,147,182],"propose":[113],"synthetic":[115,161],"approach":[117],"diversifies":[119],"amplifies":[121],"into":[124],"rich":[125],"representations,":[127],"enabling":[128],"model":[130,158],"effectively":[132],"update":[133],"its":[134,165],"parameters":[135],"limited":[137],"source":[138],"material.":[139],"reduce":[142],"reliance":[143],"data,":[146,156],"show":[148],"given":[150],"fixed":[152],"amount":[153],"such":[155],"self-generate":[160],"bootstrap":[164],"fundamental":[166],"without":[169],"distillation":[170],"any":[172],"off-the-shelf,":[173],"instruction-tuned":[174],"LM.":[175],"Finally,":[176],"transcend":[178],"human-engineered":[179],"paradigms,":[181],"demonstrate":[183],"scaling":[186],"search":[187,198],"during":[188],"test":[189],"time":[190],"over":[191,199],"space":[193,202],"algorithms,":[195],"larger":[201],"learning":[204],"algorithm":[205],"configurations":[206],"than":[207],"explore":[211],"manually.":[212]},"counts_by_year":[],"updated_date":"2026-03-21T06:36:02.116451","created_date":"2026-03-21T00:00:00"}