{"id":"https://openalex.org/W7071838162","doi":null,"title":"Transformers as Empirical Bayes Estimators The Poisson Model","display_name":"Transformers as Empirical Bayes Estimators The Poisson Model","publication_year":2025,"publication_date":"2025-02-01","ids":{"openalex":"https://openalex.org/W7071838162"},"language":"en","primary_location":{"id":"pmh:oai:dspace.mit.edu:1721.1/159119","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/1721.1/159119","pdf_url":"https://dspace.mit.edu/bitstream/1721.1/159119/-1/jabbour-mjabbour-meng-eecs-2025-thesis.pdf","source":{"id":"https://openalex.org/S4306400425","display_name":"DSpace@MIT (Massachusetts Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I63966007","host_organization_name":"Massachusetts Institute of Technology","host_organization_lineage":["https://openalex.org/I63966007"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://dspace.mit.edu/bitstream/1721.1/159119/-1/jabbour-mjabbour-meng-eecs-2025-thesis.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5116293094","display_name":"Mark Jabbour","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jabbour, Mark","raw_affiliation_strings":["MASTER OF ENGINEERING IN ELECTRICAL ENGINEERING AND COMPUTER SCIENCE","Mathematics and Computer Science and Engineering , Massachusetts Institute of Technology , 2023"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MASTER OF ENGINEERING IN ELECTRICAL ENGINEERING AND COMPUTER SCIENCE","institution_ids":[]},{"raw_affiliation_string":"Mathematics and Computer Science and Engineering , Massachusetts Institute of Technology , 2023","institution_ids":["https://openalex.org/I63966007"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5116293094"],"corresponding_institution_ids":["https://openalex.org/I63966007"],"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":true,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.36649999022483826,"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/T10028","display_name":"Topic Modeling","score":0.36649999022483826,"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":0.09309999644756317,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.09080000221729279,"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/transformer","display_name":"transformer models","score":0.9049999713897705},{"id":"https://openalex.org/keywords/empirical-bayes","display_name":"empirical Bayes","score":0.8240000009536743},{"id":"https://openalex.org/keywords/poisson-model","display_name":"Poisson model","score":0.8320000171661377},{"id":"https://openalex.org/keywords/in-context-learning","display_name":"in-context learning","score":0.9869999885559082},{"id":"https://openalex.org/keywords/maximum-likelihood-estimation","display_name":"maximum likelihood estimation","score":0.4819999933242798},{"id":"https://openalex.org/keywords/synthetic-data-pre-training","display_name":"synthetic data pre-training","score":0.5789999961853027},{"id":"https://openalex.org/keywords/intermediate-layer-representations","display_name":"intermediate layer representations","score":0.5009999871253967}],"concepts":[{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.7188000082969666},{"id":"https://openalex.org/C207201462","wikidata":"https://www.wikidata.org/wiki/Q182505","display_name":"Bayes' theorem","level":3,"score":0.6996999979019165},{"id":"https://openalex.org/C100906024","wikidata":"https://www.wikidata.org/wiki/Q205692","display_name":"Poisson distribution","level":2,"score":0.6017000079154968},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5943999886512756},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5135999917984009},{"id":"https://openalex.org/C2834757","wikidata":"https://www.wikidata.org/wiki/Q4925424","display_name":"Monotone polygon","level":2,"score":0.4575999975204468},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3799999952316284},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3467000126838684},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3337000012397766},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.3262999951839447},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.31679999828338623},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29030001163482666},{"id":"https://openalex.org/C68022304","wikidata":"https://www.wikidata.org/wiki/Q842217","display_name":"Bayes estimator","level":3,"score":0.2883000075817108},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.28279998898506165},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.274399995803833},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2727999985218048},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26649999618530273},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.2597000002861023}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:dspace.mit.edu:1721.1/159119","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/1721.1/159119","pdf_url":"https://dspace.mit.edu/bitstream/1721.1/159119/-1/jabbour-mjabbour-meng-eecs-2025-thesis.pdf","source":{"id":"https://openalex.org/S4306400425","display_name":"DSpace@MIT (Massachusetts Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I63966007","host_organization_name":"Massachusetts Institute of Technology","host_organization_lineage":["https://openalex.org/I63966007"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis"}],"best_oa_location":{"id":"pmh:oai:dspace.mit.edu:1721.1/159119","doi":null,"is_oa":true,"landing_page_url":"https://hdl.handle.net/1721.1/159119","pdf_url":"https://dspace.mit.edu/bitstream/1721.1/159119/-1/jabbour-mjabbour-meng-eecs-2025-thesis.pdf","source":{"id":"https://openalex.org/S4306400425","display_name":"DSpace@MIT (Massachusetts Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I63966007","host_organization_name":"Massachusetts Institute of Technology","host_organization_lineage":["https://openalex.org/I63966007"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Thesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7071838162.pdf","grobid_xml":"https://content.openalex.org/works/W7071838162.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W1566289585","https://openalex.org/W2626778328","https://openalex.org/W2991308675","https://openalex.org/W2995273672","https://openalex.org/W3030163527","https://openalex.org/W3198285973","https://openalex.org/W4226082499","https://openalex.org/W4296932880","https://openalex.org/W4310509152","https://openalex.org/W4313483544","https://openalex.org/W4376123079","https://openalex.org/W4383472816","https://openalex.org/W4386839891","https://openalex.org/W4391709345","https://openalex.org/W4402669802","https://openalex.org/W4413944793","https://openalex.org/W4414908202","https://openalex.org/W4415796413","https://openalex.org/W7133231369"],"related_works":["https://openalex.org/W4407632335","https://openalex.org/W4406959770","https://openalex.org/W7133209511","https://openalex.org/W4380136221","https://openalex.org/W7130410562","https://openalex.org/W4388275304","https://openalex.org/W4415795960","https://openalex.org/W4379087156","https://openalex.org/W7135131319","https://openalex.org/W4407309008"],"abstract_inverted_index":{"We":[0,110],"study":[1],"the":[2,13,19,23,28,38,41,47,51,85,89,92,101,108],"ability":[3],"of":[4,15,30,88,91,95],"transformers":[5,31,57,116],"to":[6,36,64],"perform":[7],"In":[8],"Context":[9],"Learning":[10],"(ICL)":[11],"in":[12,79,123],"setting":[14],"Empirical":[16],"Bayes":[17,44],"for":[18,46],"Poison":[20],"Model.":[21],"On":[22,50],"theoretical":[24],"side,":[25,53],"we":[26,54,98],"demonstrate":[27,112],"expressibility":[29],"by":[32],"formulating":[33],"a":[34],"way":[35],"approximate":[37],"Robbins":[39],"estimator,":[40],"first":[42],"empirical":[43,52],"estimator":[45],"Poisson":[48],"model.":[49],"show":[55],"that":[56,100,113],"pre-trained":[58],"on":[59],"synthetic":[60],"data":[61],"can":[62],"generalize":[63],"unseen":[65],"prior":[66],"and":[67,76,81,105],"sequence":[68],"lengths,":[69],"outperforming":[70],"existing":[71],"methods":[72],"like":[73],"Robbins,":[74],"NPMLE,":[75],"ERM":[77],"monotone":[78],"efficiency":[80],"accuracy.":[82],"By":[83],"studying":[84],"internal":[86],"behavior":[87],"representations":[90],"intermediate":[93],"layers":[94],"these":[96],"transformers,":[97],"found":[99],"representation":[102],"converges":[103],"quickly":[104],"smoothly":[106],"over":[107],"layers.":[109],"also":[111],"it\u2019s":[114],"unlikely":[115],"are":[117],"implementing":[118],"Robbin\u2019s":[119],"or":[120],"NPMLE":[121],"estimators":[122],"context.":[124]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}