{"id":"https://openalex.org/W7223239297","doi":"https://doi.org/10.48550/arxiv.2610.28977","title":"Is In-Context Reinforcement Learning A Posterior Sampler? An Empirical Study","display_name":"Is In-Context Reinforcement Learning A Posterior Sampler? An Empirical Study","publication_year":2026,"publication_date":"2026-10-20","ids":{"openalex":"https://openalex.org/W7223239297","doi":"https://doi.org/10.48550/arxiv.2610.28977"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.28977","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.28977","pdf_url":"https://arxiv.org/pdf/2610.28977","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/2610.28977","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5123069484","display_name":"Weiqin Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen, Weiqin","raw_affiliation_strings":["School of Industrial Engineering , Purdue University West Lafayette , IN 47907 , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Industrial Engineering , Purdue University West Lafayette , IN 47907 , USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5159276118","display_name":"Xinlong Du","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Du, Xinlong","raw_affiliation_strings":["School of Industrial Engineering , Purdue University West Lafayette , IN 47907 , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Industrial Engineering , Purdue University West Lafayette , IN 47907 , USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041765559","display_name":"Gesualdo Scutari","orcid":"https://orcid.org/0000-0002-6453-6870","observed_orcids":["https://orcid.org/0000-0002-6453-6870"]},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Scutari, Gesualdo","raw_affiliation_strings":["School of Industrial Engineering , Purdue University West Lafayette , IN 47907 , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Industrial Engineering , Purdue University West Lafayette , IN 47907 , USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5070127492","display_name":"Harsha Honnappa","orcid":"https://orcid.org/0000-0002-0834-054X","observed_orcids":["https://orcid.org/0000-0002-0834-054X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Honnappa, Harsha","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.6923999786376953,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.6923999786376953,"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.23520000278949738,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.015300000086426735,"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/bayesian-posterior-sampling","display_name":"Bayesian posterior sampling","score":0.9779999852180481},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7929999828338623},{"id":"https://openalex.org/keywords/martingale-property","display_name":"martingale property","score":1.0},{"id":"https://openalex.org/keywords/zero-shot-generalization","display_name":"zero-shot generalization","score":0.6819999814033508},{"id":"https://openalex.org/keywords/bandit-problems","display_name":"bandit problems","score":0.5870000123977661}],"concepts":[{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.677299976348877},{"id":"https://openalex.org/C57830394","wikidata":"https://www.wikidata.org/wiki/Q278079","display_name":"Posterior probability","level":3,"score":0.6674000024795532},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.599399983882904},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.5706999897956848},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5662000179290771},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5304999947547913},{"id":"https://openalex.org/C101112237","wikidata":"https://www.wikidata.org/wiki/Q4874481","display_name":"Bayesian statistics","level":4,"score":0.5157999992370605},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.5037999749183655},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4909999966621399},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.43810001015663147},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.43369999527931213},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.42739999294281006},{"id":"https://openalex.org/C207201462","wikidata":"https://www.wikidata.org/wiki/Q182505","display_name":"Bayes' theorem","level":3,"score":0.4156999886035919},{"id":"https://openalex.org/C99173435","wikidata":"https://www.wikidata.org/wiki/Q4874469","display_name":"Bayesian experimental design","level":5,"score":0.41429999470710754},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4052000045776367},{"id":"https://openalex.org/C917703","wikidata":"https://www.wikidata.org/wiki/Q7239668","display_name":"Predictive inference","level":5,"score":0.3912999927997589},{"id":"https://openalex.org/C97933134","wikidata":"https://www.wikidata.org/wiki/Q5374249","display_name":"Empirical probability","level":4,"score":0.3476000130176544},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3287999927997589},{"id":"https://openalex.org/C37903108","wikidata":"https://www.wikidata.org/wiki/Q4874474","display_name":"Bayesian linear regression","level":4,"score":0.3278000056743622},{"id":"https://openalex.org/C142291917","wikidata":"https://www.wikidata.org/wiki/Q4165283","display_name":"Bayes factor","level":4,"score":0.3239000141620636},{"id":"https://openalex.org/C71983512","wikidata":"https://www.wikidata.org/wiki/Q7915687","display_name":"Variable-order Bayesian network","level":4,"score":0.296099990606308},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.28299999237060547},{"id":"https://openalex.org/C68022304","wikidata":"https://www.wikidata.org/wiki/Q842217","display_name":"Bayes estimator","level":3,"score":0.2793999910354614},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C80478641","wikidata":"https://www.wikidata.org/wiki/Q195771","display_name":"Sequential analysis","level":2,"score":0.26159998774528503},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.2531999945640564},{"id":"https://openalex.org/C149569020","wikidata":"https://www.wikidata.org/wiki/Q25098598","display_name":"Bayesian average","level":5,"score":0.2524999976158142},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.28977","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.28977","pdf_url":"https://arxiv.org/pdf/2610.28977","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.2610.28977","doi":"https://doi.org/10.48550/arxiv.2610.28977","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.28977","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:2610.28977","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.28977","pdf_url":"https://arxiv.org/pdf/2610.28977","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":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7223239297.pdf","grobid_xml":"https://content.openalex.org/works/W7223239297.grobid-xml"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W1757796397","https://openalex.org/W2736601468","https://openalex.org/W3123348991","https://openalex.org/W4323697341","https://openalex.org/W4327810158","https://openalex.org/W4379527413","https://openalex.org/W4384918448","https://openalex.org/W4387561528","https://openalex.org/W4387635993","https://openalex.org/W7090846667"],"related_works":["https://openalex.org/W4379087156","https://openalex.org/W4406959770","https://openalex.org/W4399399219","https://openalex.org/W7090846667","https://openalex.org/W4380136221","https://openalex.org/W7162428732","https://openalex.org/W4405253784","https://openalex.org/W7119169134","https://openalex.org/W7133242219","https://openalex.org/W4387635993"],"abstract_inverted_index":{"In-context":[0],"reinforcement":[1],"learning":[2,8],"(ICRL)":[3],"exploits":[4],"the":[5,31,106,167],"impressive":[6],"in-context":[7],"ability":[9],"of":[10,170],"large":[11],"language":[12],"models":[13,121],"to":[14,17,115],"zero-shot":[15],"generalize":[16],"new,":[18],"unseen":[19],"tasks":[20],"at":[21],"inference":[22],"time":[23],"without":[24],"any":[25],"parameter":[26],"updates.":[27],"Among":[28],"many":[29],"approaches,":[30],"Decision-Pretrained":[32],"Transformer":[33],"(DPT)":[34],"is":[35],"particularly":[36],"appealing":[37],"for":[38,165],"its":[39],"theoretical":[40],"grounding:":[41],"under":[42],"idealized":[43],"assumptions":[44],"such":[45],"as":[46,149],"a":[47,81,97,162],"sufficiently":[48],"expressive":[49],"model":[50],"and":[51,76,105,131],"infinite":[52],"pretraining":[53],"data,":[54],"DPT":[55,75,79],"can":[56],"perform":[57],"Bayesian":[58,82,98,150,168],"posterior":[59,83,99,151],"sampling.":[60],"Whether":[61],"this":[62,89,156],"characterization":[63],"survives":[64],"in":[65,85,128,153],"practice":[66],"remains":[67],"open.":[68],"Therefore,":[69],"we":[70,91,135],"ask:":[71],"\\textit{is":[72],"ICRL,":[73],"specifically":[74],"our":[77,159],"proposed":[78],"variants,":[80],"sampler":[84,100],"practice?}":[86],"To":[87],"approach":[88],"question,":[90],"establish":[92],"two":[93],"necessary":[94],"conditions":[95,139],"that":[96,137,144],"must":[101],"satisfy:":[102],"\\textit{sufficient-statistic":[103],"invariance}":[104],"\\textit{martingale":[107],"property}.":[108],"We":[109],"then":[110],"construct":[111],"statistical":[112],"hypothesis":[113],"tests":[114,127,160],"statistically":[116],"assess":[117],"if":[118],"practically-trained":[119],"ICRL":[120,145],"satisfy":[122],"these":[123,126],"conditions.":[124],"Applying":[125],"Gaussian,":[129],"Bernoulli,":[130],"linear":[132],"bandit":[133],"settings,":[134],"find":[136],"both":[138],"are":[140],"consistently":[141],"violated,":[142],"suggesting":[143],"may":[146],"not":[147],"behave":[148],"samplers":[152],"practice.":[154],"Beyond":[155],"negative":[157],"result,":[158],"provide":[161],"principled":[163],"toolkit":[164],"testing":[166],"behavior":[169],"ICRL.":[171]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-23T00:00:00"}