{"id":"https://openalex.org/W7163587735","doi":"https://doi.org/10.48550/arxiv.2606.04703","title":"Rethinking Continual Experience Internalization for Self-Evolving LLM Agents","display_name":"Rethinking Continual Experience Internalization for Self-Evolving LLM Agents","publication_year":2026,"publication_date":"2026-06-03","ids":{"openalex":"https://openalex.org/W7163587735","doi":"https://doi.org/10.48550/arxiv.2606.04703"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.04703","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.04703","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.04703","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137917743","display_name":"Jingwen Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jingwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137814592","display_name":"Wenkai Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Wenkai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112974345","display_name":"Shengda Fan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fan, Shengda","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030688626","display_name":"Wenbo Nie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nie, Wenbo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137875335","display_name":"Chenxing Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Chenxing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137890557","display_name":"Shaodong Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Shaodong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110942658","display_name":"Yunlei Hu","orcid":"https://orcid.org/0009-0001-2746-4935"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Yangen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137917617","display_name":"Lu Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137900319","display_name":"Ke Zeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Ke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137856433","display_name":"Yankai Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Yankai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.3098999857902527,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.3098999857902527,"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":0.16179999709129333,"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/T10636","display_name":"Innovative Teaching and Learning Methods","score":0.040699999779462814,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/internalization","display_name":"Internalization","score":0.5860999822616577},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.5148000121116638},{"id":"https://openalex.org/keywords/experiential-learning","display_name":"Experiential learning","score":0.4171999990940094},{"id":"https://openalex.org/keywords/yield","display_name":"Yield (engineering)","score":0.3546999990940094},{"id":"https://openalex.org/keywords/property","display_name":"Property (philosophy)","score":0.350600004196167},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.3481999933719635},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.3481999933719635}],"concepts":[{"id":"https://openalex.org/C139770010","wikidata":"https://www.wikidata.org/wiki/Q1339807","display_name":"Internalization","level":3,"score":0.5860999822616577},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.5618000030517578},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.5148000121116638},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4571000039577484},{"id":"https://openalex.org/C37228920","wikidata":"https://www.wikidata.org/wiki/Q1307600","display_name":"Experiential learning","level":2,"score":0.4171999990940094},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.3921000063419342},{"id":"https://openalex.org/C195094911","wikidata":"https://www.wikidata.org/wiki/Q14167904","display_name":"Process management","level":1,"score":0.36899998784065247},{"id":"https://openalex.org/C134121241","wikidata":"https://www.wikidata.org/wiki/Q899301","display_name":"Yield (engineering)","level":2,"score":0.3546999990940094},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.350600004196167},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.34929999709129333},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.3481999933719635},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.3481999933719635},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.3368000090122223},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.33480000495910645},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3310000002384186},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.32409998774528503},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.32010000944137573},{"id":"https://openalex.org/C34585555","wikidata":"https://www.wikidata.org/wiki/Q1368723","display_name":"Learning curve","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.30640000104904175},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.3001999855041504},{"id":"https://openalex.org/C55587333","wikidata":"https://www.wikidata.org/wiki/Q1133029","display_name":"Engineering ethics","level":1,"score":0.299699991941452},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.28110000491142273},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C184356942","wikidata":"https://www.wikidata.org/wiki/Q830382","display_name":"Best practice","level":2,"score":0.26759999990463257},{"id":"https://openalex.org/C66204764","wikidata":"https://www.wikidata.org/wiki/Q219416","display_name":"Sustainability","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.04703","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.04703","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.04703","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.04703","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.42257627844810486,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Experience":[0,65,89],"internalization":[1],"converts":[2],"contextual":[3],"experience":[4,38,62,71,104,164],"from":[5,43,85],"past":[6],"interactions":[7],"into":[8],"reusable":[9],"parametric":[10],"capability,":[11],"offering":[12],"a":[13,44,109,131,155],"promising":[14],"path":[15],"toward":[16],"continual":[17],"learning":[18,174],"in":[19],"large":[20],"language":[21],"models":[22],"(LLMs).":[23],"While":[24],"prior":[25],"work":[26],"has":[27],"predominantly":[28],"focused":[29],"on":[30,126,147],"single-iteration":[31],"transfer,":[32],"we":[33],"discover":[34],"that":[35,69,95,111,123],"under":[36],"multi-iteration":[37],"learning,":[39],"existing":[40],"methods":[41],"suffer":[42],"progressive":[45],"capability":[46],"collapse":[47],"rather":[48],"than":[49,75,137],"compounding":[50],"improvement.":[51],"We":[52,67,121],"systematically":[53],"examine":[54],"this":[55],"failure":[56],"through":[57],"three":[58],"vital":[59],"dimensions":[60],"of":[61],"internalization:":[63],"(1)":[64],"Granularity:":[66],"find":[68],"principle-level":[70],"is":[72,112,141],"more":[73,133],"durable":[74],"instance-level":[76],"experience,":[77],"as":[78],"it":[79],"effectively":[80],"abstracts":[81],"transferable":[82],"strategies":[83],"away":[84],"trajectory-specific":[86],"details.":[87],"(2)":[88],"Injection":[90],"Pattern:":[91],"Our":[92],"analysis":[93],"reveals":[94],"step-wise":[96],"injection":[97,101],"significantly":[98],"outperforms":[99],"global":[100],"by":[102,144],"aligning":[103],"with":[105],"intermediate":[106],"decision":[107],"states,":[108],"property":[110],"critical":[113],"for":[114,160,169],"long-horizon":[115],"tool":[116],"use.":[117],"(3)":[118],"Internalization":[119],"Regime:":[120],"demonstrate":[122],"off-policy":[124],"context-distillation":[125],"high-quality":[127],"teacher":[128],"trajectories":[129],"provides":[130],"substantially":[132],"stable":[134,161],"training":[135],"signal":[136],"on-policy":[138],"context-distillation,":[139],"which":[140],"inherently":[142],"limited":[143],"local":[145],"corrections":[146],"student-induced":[148],"flawed":[149],"states.":[150],"Together,":[151],"these":[152],"insights":[153],"yield":[154],"simple":[156],"yet":[157],"robust":[158],"recipe":[159],"and":[162,172],"sustainable":[163],"internalization,":[165],"providing":[166],"concrete":[167],"guidance":[168],"engineering":[170],"self-evolving":[171],"continually":[173],"LLMs.":[175]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-05T00:00:00"}