{"id":"https://openalex.org/W7162557508","doi":"https://doi.org/10.48550/arxiv.2605.26713","title":"Transformers Can Learn Posterior Predictive Distributions In-Context","display_name":"Transformers Can Learn Posterior Predictive Distributions In-Context","publication_year":2026,"publication_date":"2026-05-26","ids":{"openalex":"https://openalex.org/W7162557508","doi":"https://doi.org/10.48550/arxiv.2605.26713"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.26713","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.26713","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.26713","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5074775363","display_name":"Gyeonghun Kang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kang, Gyeonghun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023256126","display_name":"Changwoo J. Lee","orcid":"https://orcid.org/0000-0003-2132-148X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Changwoo J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137110321","display_name":"Xiang Cheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cheng, Xiang","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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.8614000082015991,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.8614000082015991,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.03350000083446503,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.01269999984651804,"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/posterior-probability","display_name":"Posterior probability","score":0.6054999828338623},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.5149999856948853},{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.4749999940395355},{"id":"https://openalex.org/keywords/extrapolation","display_name":"Extrapolation","score":0.4323999881744385},{"id":"https://openalex.org/keywords/gaussian-process","display_name":"Gaussian process","score":0.4307999908924103},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.42809998989105225},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.42149999737739563},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.42100000381469727}],"concepts":[{"id":"https://openalex.org/C57830394","wikidata":"https://www.wikidata.org/wiki/Q278079","display_name":"Posterior probability","level":3,"score":0.6054999828338623},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.5149999856948853},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5090000033378601},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.4749999940395355},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.44859999418258667},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4339999854564667},{"id":"https://openalex.org/C132459708","wikidata":"https://www.wikidata.org/wiki/Q744069","display_name":"Extrapolation","level":2,"score":0.4323999881744385},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.4307999908924103},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.42809998989105225},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.42149999737739563},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.42100000381469727},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3693999946117401},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35100001096725464},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.3174000084400177},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.3043999969959259},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2994999885559082},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.29510000348091125},{"id":"https://openalex.org/C41426520","wikidata":"https://www.wikidata.org/wiki/Q1192065","display_name":"Point estimation","level":2,"score":0.2892000079154968},{"id":"https://openalex.org/C70153297","wikidata":"https://www.wikidata.org/wiki/Q5591907","display_name":"Gradient boosting","level":3,"score":0.2603999972343445},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.25450000166893005}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.26713","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.26713","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.26713","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.26713","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":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Prior-data":[0],"fitted":[1],"networks":[2],"(PFNs)":[3],"have":[4],"recently":[5],"emerged":[6],"as":[7],"a":[8,66],"powerful":[9],"approach":[10],"for":[11],"Bayesian":[12],"prediction":[13],"tasks,":[14],"approximating":[15],"the":[16,39,71,89,93,111,117,124,130,145],"posterior":[17,72],"predictive":[18,73],"distribution":[19],"(PPD)":[20],"through":[21],"in-context":[22],"learning.":[23],"Despite":[24],"their":[25],"strong":[26],"empirical":[27],"performance":[28],"and":[29,75,101,116,151],"ability":[30],"to":[31,44],"go":[32],"beyond":[33,129],"point":[34],"predictions,":[35],"theoretical":[36],"understandings":[37],"of":[38,42,85,92,98,114,119,127,147],"algorithmic":[40],"capability":[41],"transformers":[43,63,128],"learn":[45],"distributions":[46],"in":[47,96,122],"context":[48],"are":[49],"still":[50],"lacking.":[51],"Focusing":[52],"on":[53,105],"Gaussian":[54],"process":[55],"regression":[56],"problems,":[57],"we":[58,108],"show":[59],"by":[60,78],"construction":[61],"that":[62,81,138],"can":[64],"implement":[65],"gradient":[67],"descent":[68],"algorithm":[69],"targeting":[70,149],"mean":[74],"variance,":[76],"followed":[77],"nonlinear":[79],"mappings":[80],"yield":[82],"binned":[83],"probabilities":[84],"PPD.":[86],"We":[87,135],"study":[88],"error":[90],"bounds":[91],"approximated":[94],"PPD":[95],"terms":[97],"attention":[99,120],"depth":[100,121],"bin":[102],"resolution.":[103],"Based":[104],"these":[106],"results,":[107],"further":[109],"demonstrate":[110],"key":[112],"role":[113],"normalization":[115],"choice":[118],"enabling":[123],"extrapolation":[125],"abilities":[126],"pretraining":[131],"sample":[132],"size":[133],"range.":[134],"conduct":[136],"simulations":[137],"corroborate":[139],"our":[140],"findings,":[141],"providing":[142],"insight":[143],"into":[144],"expressivity":[146],"PFNs":[148],"PPDs":[150],"how":[152],"architectural":[153],"choices":[154],"may":[155],"influence":[156],"generalization":[157],"capabilities.":[158]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-28T00:00:00"}