{"id":"https://openalex.org/W7214909346","doi":"https://doi.org/10.48550/arxiv.2609.35911","title":"Learn Now, Use Next, Trust Later: Prequential Test-Time Learning for LLM Agents","display_name":"Learn Now, Use Next, Trust Later: Prequential Test-Time Learning for LLM Agents","publication_year":2026,"publication_date":"2026-09-28","ids":{"openalex":"https://openalex.org/W7214909346","doi":"https://doi.org/10.48550/arxiv.2609.35911"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.35911","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.35911","pdf_url":"https://arxiv.org/pdf/2609.35911","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/2609.35911","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153631991","display_name":"Tong Zhao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhao, Tong","raw_affiliation_strings":["Renmin University of China ;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Renmin University of China ;","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5146575019","display_name":"Reed Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li, Reed","raw_affiliation_strings":["Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153603623","display_name":"Yuyang Hu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hu, Yuyang","raw_affiliation_strings":["Renmin University of China ;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Renmin University of China ;","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007186741","display_name":"Yutao Zhu","orcid":"https://orcid.org/0000-0002-9432-3251","observed_orcids":["https://orcid.org/0000-0002-9432-3251"]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhu, Yutao","raw_affiliation_strings":["Renmin University of China ;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Renmin University of China ;","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048582949","display_name":"Haijin Liang","orcid":"https://orcid.org/0009-0006-3464-9192","observed_orcids":["https://orcid.org/0009-0006-3464-9192"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang, Haijin","raw_affiliation_strings":["Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153609327","display_name":"Haibo Shi","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shi, Haibo","raw_affiliation_strings":["Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153684189","display_name":"Yu Lu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lu, Yu","raw_affiliation_strings":["Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5153611452","display_name":"Zhicheng Dou","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dou, Zhicheng","raw_affiliation_strings":["Renmin University of China ;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Renmin University of China ;","institution_ids":["https://openalex.org/I78988378"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":0.7246000170707703,"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":0.7246000170707703,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.12240000069141388,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.06920000165700912,"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/llm-based-agents","display_name":"LLM agents","score":0.9610000252723694},{"id":"https://openalex.org/keywords/nonparametric-learning","display_name":"nonparametric learning","score":0.5989999771118164},{"id":"https://openalex.org/keywords/hypothesis-validation","display_name":"hypothesis verification","score":0.4519999921321869},{"id":"https://openalex.org/keywords/knowledge-reuse","display_name":"knowledge reuse","score":0.4059999883174896},{"id":"https://openalex.org/keywords/alfworld","display_name":"ALFWorld","score":0.9950000047683716}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7099999785423279},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.6929000020027161},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5565999746322632},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.531499981880188},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4921000003814697},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4244999885559082},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.40700000524520874},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3799999952316284},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.3637000024318695},{"id":"https://openalex.org/C108650721","wikidata":"https://www.wikidata.org/wiki/Q1783253","display_name":"Counterfactual thinking","level":2,"score":0.33329999446868896},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.32519999146461487},{"id":"https://openalex.org/C102366305","wikidata":"https://www.wikidata.org/wiki/Q1097688","display_name":"Nonparametric statistics","level":2,"score":0.32359999418258667},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.32269999384880066},{"id":"https://openalex.org/C39920170","wikidata":"https://www.wikidata.org/wiki/Q693083","display_name":"Soundness","level":2,"score":0.3212999999523163},{"id":"https://openalex.org/C34585555","wikidata":"https://www.wikidata.org/wiki/Q1368723","display_name":"Learning curve","level":2,"score":0.31200000643730164},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.295199990272522},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2685000002384186},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2671000063419342},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.2605000138282776}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.35911","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.35911","pdf_url":"https://arxiv.org/pdf/2609.35911","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.2609.35911","doi":"https://doi.org/10.48550/arxiv.2609.35911","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.35911","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:2609.35911","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.35911","pdf_url":"https://arxiv.org/pdf/2609.35911","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214909346.pdf","grobid_xml":"https://content.openalex.org/works/W7214909346.grobid-xml"},"referenced_works_count":13,"referenced_works":["https://openalex.org/W4387636003","https://openalex.org/W4387835442","https://openalex.org/W4393147158","https://openalex.org/W4393160747","https://openalex.org/W4414384539","https://openalex.org/W4415337833","https://openalex.org/W4417248776","https://openalex.org/W7125787811","https://openalex.org/W7140107139","https://openalex.org/W7161705332","https://openalex.org/W7162490112","https://openalex.org/W7163637351","https://openalex.org/W7167209728"],"related_works":["https://openalex.org/W7203875578","https://openalex.org/W7167570570","https://openalex.org/W7220734018","https://openalex.org/W7160858818","https://openalex.org/W7168354618","https://openalex.org/W4412888257","https://openalex.org/W7161058656","https://openalex.org/W7166876537","https://openalex.org/W7163990747","https://openalex.org/W7166813837"],"abstract_inverted_index":{"Adapting":[0],"large":[1],"language":[2],"model":[3,163],"agents":[4],"during":[5],"deployment":[6],"requires":[7],"not":[8,38,76,213],"only":[9,146],"retaining":[10],"past":[11],"experience,":[12],"but":[13,62],"also":[14],"turning":[15],"new":[16],"observations":[17,140],"into":[18,41,117],"timely":[19],"guidance.":[20,99,154],"Many":[21],"test-time":[22],"learning":[23],"methods,":[24],"however,":[25],"acquire":[26],"knowledge":[27,42,51,159],"from":[28,32,110],"completed":[29],"episodes.":[30,83,132],"Feedback":[31],"an":[33],"ongoing":[34],"interaction":[35],"may":[36,75],"therefore":[37],"be":[39,77],"distilled":[40],"soon":[43],"enough":[44,79],"to":[45,80,215],"help":[46],"the":[47,53,122,142,194,202,216],"next":[48,123],"decision.":[49],"Acquiring":[50],"at":[52],"granularity":[54],"of":[55,178],"individual":[56],"transitions":[57,116],"could":[58],"reduce":[59],"this":[60],"delay,":[61],"raises":[63],"a":[64,67,103],"separate":[65],"challenge:":[66],"rule":[68],"that":[69,106,119,209],"is":[70],"useful":[71],"within":[72],"one":[73],"episode":[74],"reliable":[78],"guide":[81,121],"future":[82],"Waiting":[84],"for":[85,152],"validation":[86,128],"can":[87,94,120],"forfeit":[88],"immediate":[89,108],"benefits,":[90],"whereas":[91],"unrestricted":[92],"reuse":[93,130],"propagate":[95],"accidental":[96],"or":[97],"misattributed":[98],"We":[100],"introduce":[101],"StepLearn,":[102],"nonparametric":[104],"framework":[105],"separates":[107],"use":[109],"persistent":[111,153],"trust.":[112],"It":[113,191],"turns":[114],"informative":[115],"hypotheses":[118,149],"step,":[124],"while":[125,160],"requiring":[126],"prospective":[127],"before":[129],"across":[131,201],"Their":[133],"predicted":[134],"effects":[135],"are":[136,212],"checked":[137],"against":[138],"subsequent":[139],"outside":[141],"source":[143],"episodes,":[144],"and":[145,171,180,184,186],"sufficiently":[147],"supported":[148],"become":[150],"available":[151],"This":[155],"process":[156],"updates":[157],"external":[158],"keeping":[161],"all":[162],"parameters":[164],"fixed.":[165],"Over":[166],"five":[167],"rounds":[168],"on":[169,223],"WebArena-Lite":[170],"ALFWorld,":[172],"StepLearn":[173],"achieves":[174],"average":[175],"success":[176],"rates":[177],"59.9%":[179],"84.0%":[181],"with":[182,188,219],"GPT-5-mini,":[183],"57.8%":[185],"88.1%":[187],"Qwen3.5-35B-A3B,":[189],"respectively.":[190],"outperforms":[192],"EvoTest,":[193],"strongest":[195],"baseline,":[196],"by":[197],"2.2-12.7":[198],"percentage":[199],"points":[200],"four":[203],"settings.":[204,229],"Learning":[205],"dynamics":[206],"further":[207],"shows":[208],"these":[210],"gains":[211],"restricted":[214],"final":[217],"repetition,":[218],"advantages":[220],"already":[221],"present":[222],"first":[224],"task":[225],"attempts":[226],"in":[227],"most":[228]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-01T00:00:00"}