{"id":"https://openalex.org/W7160595679","doi":"https://doi.org/10.48550/arxiv.2605.05413","title":"From History to State: Constant-Context Skill Learning for LLM Agents","display_name":"From History to State: Constant-Context Skill Learning for LLM Agents","publication_year":2026,"publication_date":"2026-05-06","ids":{"openalex":"https://openalex.org/W7160595679","doi":"https://doi.org/10.48550/arxiv.2605.05413"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.05413","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05413","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.2605.05413","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135656964","display_name":"Haoyang Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Haoyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135679787","display_name":"Xinyuan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xinyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135669592","display_name":"Yancheng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yancheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135655780","display_name":"Puda Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Puda","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135721141","display_name":"Feng Ju","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ju, Feng","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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.11959999799728394,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.11959999799728394,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.09319999814033508,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.07000000029802322,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6456000208854675},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.5562000274658203},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5235999822616577},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4864000082015991},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.45669999718666077},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.4066999852657318},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.36629998683929443},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.33709999918937683},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.3327000141143799}],"concepts":[{"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/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6456000208854675},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.5562000274658203},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5235999822616577},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.49720001220703125},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4864000082015991},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.45669999718666077},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4066999852657318},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3822999894618988},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.36629998683929443},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.33709999918937683},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.3327000141143799},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C2777620637","wikidata":"https://www.wikidata.org/wiki/Q7535191","display_name":"Skill mix","level":3,"score":0.30820000171661377},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.30090001225471497},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2856999933719635},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.2842999994754791},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.2818000018596649},{"id":"https://openalex.org/C132758656","wikidata":"https://www.wikidata.org/wiki/Q5307365","display_name":"Dreyfus model of skill acquisition","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.2777999937534332},{"id":"https://openalex.org/C2776608160","wikidata":"https://www.wikidata.org/wiki/Q4785462","display_name":"Natural (archaeology)","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.2628999948501587},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C74072328","wikidata":"https://www.wikidata.org/wiki/Q1142726","display_name":"Intelligent agent","level":2,"score":0.25619998574256897},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.25429999828338623},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.25279998779296875}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.05413","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05413","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.2605.05413","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05413","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"model":[2],"(LLM)":[3],"agents":[4,24,132],"are":[5,78],"increasingly":[6],"used":[7],"to":[8,40,177],"operate":[9],"browsers,":[10],"files,":[11],"code":[12],"and":[13,61,92,107,121,129,139,154],"tools,":[14],"making":[15],"personal":[16,23],"assistants":[17],"a":[18,26,69,93],"natural":[19],"deployment":[20],"target.":[21],"Yet":[22],"face":[25],"privacy-cost-capability":[27],"tension:":[28],"cloud":[29],"models":[30,45],"execute":[31],"multi-step":[32],"workflows":[33],"well":[34],"but":[35,48],"expose":[36],"sensitive":[37],"intermediate":[38],"context":[39,185],"external":[41],"APIs,":[42],"while":[43,84,168],"local":[44],"preserve":[46],"privacy":[47],"remain":[49],"less":[50],"reliable.":[51],"Both":[52],"settings":[53],"also":[54],"pay":[55],"repeatedly":[56],"for":[57,72],"long":[58],"skill":[59,67],"prompts":[60,190],"growing":[62],"histories.":[63],"We":[64],"propose":[65],"constant-context":[66],"learning,":[68],"context-to-weights":[70],"framework":[71],"recurring":[73],"agent":[74],"workflows:":[75],"reusable":[76],"procedures":[77],"learned":[79],"in":[80],"lightweight":[81],"task-family":[82],"modules,":[83],"inference":[85],"conditions":[86],"only":[87],"on":[88,148,152,158],"the":[89],"current":[90],"observation":[91],"compact":[94],"state":[95,102],"block.":[96],"A":[97],"deterministic":[98],"tracker":[99],"renders":[100],"this":[101],"block":[103],"from":[104,189],"task":[105],"progress":[106],"supplies":[108],"aligned":[109],"subgoal":[110],"rewards,":[111],"so":[112],"each":[113],"module":[114],"can":[115,186],"be":[116,187],"trained":[117],"with":[118],"step-level":[119],"SFT":[120],"refined":[122],"through":[123],"online":[124],"RL.":[125],"Across":[126],"ALFWorld,":[127,149],"WebShop,":[128,153],"SciWorld,":[130],"our":[131],"achieve":[133],"strong":[134,164],"performance":[135],"across":[136],"Qwen3-4B,":[137],"Qwen3-8B":[138],"Llama-3.1-8B.":[140],"With":[141],"Qwen3-8B,":[142],"SFT+RL":[143],"reaches":[144],"89.6\\%":[145],"unseen":[146,156],"success":[147,151,157],"76.8\\%":[150],"66.4\\%":[155],"SciWorld.":[159],"They":[160],"match":[161],"or":[162],"exceed":[163],"published":[165],"agent-training":[166],"results":[167],"reducing":[169],"prompt":[170],"tokens":[171],"per":[172],"turn":[173],"by":[174],"2--7$\\times$":[175],"relative":[176],"controlled":[178],"ReAct":[179],"prompting":[180],"baselines,":[181],"showing":[182],"that":[183],"procedural":[184],"moved":[188],"into":[191],"weights.":[192]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-09T00:00:00"}