{"id":"https://openalex.org/W7155885443","doi":"https://doi.org/10.48550/arxiv.2604.22678","title":"BERAG: Bayesian Ensemble Retrieval-Augmented Generation for Knowledge-based Visual Question Answering","display_name":"BERAG: Bayesian Ensemble Retrieval-Augmented Generation for Knowledge-based Visual Question Answering","publication_year":2026,"publication_date":"2026-04-24","ids":{"openalex":"https://openalex.org/W7155885443","doi":"https://doi.org/10.48550/arxiv.2604.22678"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.22678","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.22678","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.22678","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134664836","display_name":"Jinghong Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jinghong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071561232","display_name":"Jingbiao Mei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mei, Jingbiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103753392","display_name":"Guangyu Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Guangyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134698745","display_name":"Bill Byrne","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Byrne, Bill","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9939000010490417,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9939000010490417,"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/T10028","display_name":"Topic Modeling","score":0.002099999925121665,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.0010999999940395355,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.7124999761581421},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.6118999719619751},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5227000117301941},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4740000069141388},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.4731999933719635},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.4348999857902527},{"id":"https://openalex.org/keywords/bayesian-inference","display_name":"Bayesian inference","score":0.3978999853134155},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.39079999923706055}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7853999733924866},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.7124999761581421},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.6118999719619751},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5338000059127808},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5227000117301941},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4740000069141388},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.4731999933719635},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.4348999857902527},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4002000093460083},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.3978999853134155},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3919000029563904},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.39079999923706055},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.3885999917984009},{"id":"https://openalex.org/C87619178","wikidata":"https://www.wikidata.org/wiki/Q126002","display_name":"Concatenation (mathematics)","level":2,"score":0.36640000343322754},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.3328999876976013},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.329800009727478},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3005000054836273},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.28540000319480896},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C159423971","wikidata":"https://www.wikidata.org/wiki/Q177251","display_name":"Associative property","level":2,"score":0.26339998841285706},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.2603999972343445},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.25220000743865967}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.22678","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.22678","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.22678","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.22678","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":"article"},"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":{"A":[0],"common":[1],"approach":[2,164],"to":[3,11,21,25,45,86,240],"question":[4,83,193],"answering":[5,194],"with":[6,65,119],"retrieval-augmented":[7],"generation":[8],"(RAG)":[9],"is":[10,55],"concatenate":[12],"documents":[13,137],"into":[14],"a":[15,22,68,125,140],"single":[16,141],"context":[17,66,76,92],"and":[18,43,152,171,187,221,244],"pass":[19],"it":[20,178],"language":[23,130],"model":[24],"generate":[26],"an":[27],"answer.":[28],"While":[29],"simple,":[30],"this":[31],"strategy":[32],"can":[33,94,237],"obscure":[34],"the":[35,46,75,104,231],"contribution":[36],"of":[37,174],"individual":[38,135],"documents,":[39],"making":[40,177],"attribution":[41,173],"difficult":[42],"contributing":[44],"``lost-in-the-middle''":[47,232],"effect,":[48],"where":[49,196],"relevant":[50],"information":[51],"in":[52,81,128],"long":[53],"contexts":[54],"overlooked.":[56],"Concatenation":[57],"also":[58,226],"scales":[59],"poorly:":[60],"computational":[61],"cost":[62],"grows":[63],"quadratically":[64],"length,":[67],"problem":[69],"that":[70,228],"becomes":[71],"especially":[72],"severe":[73],"when":[74],"includes":[77],"visual":[78,82,192],"data,":[79],"as":[80,124,149],"answering.":[84],"Attempts":[85],"mitigate":[87],"these":[88],"issues":[89],"by":[90,98,108,156],"limiting":[91],"length":[93],"further":[95],"restrict":[96],"performance":[97],"preventing":[99],"models":[100,131,197],"from":[101,103],"benefiting":[102],"improved":[105],"recall":[106],"offered":[107],"deeper":[109],"retrieval.":[110],"We":[111,184,225],"propose":[112],"Bayesian":[113,120],"Ensemble":[114,121],"Retrieval-Augmented":[115],"Generation":[116],"(BERAG),":[117],"along":[118],"Fine-Tuning":[122],"(BEFT),":[123],"RAG":[126],"framework":[127],"which":[129],"are":[132],"conditioned":[133],"on":[134,190,216],"retrieved":[136],"rather":[138],"than":[139,253],"combined":[142],"context.":[143],"BERAG":[144,186,229],"treats":[145],"document":[146,175,182,235,248],"posterior":[147,236],"probabilities":[148],"ensemble":[150],"weights":[151],"updates":[153],"them":[154],"token":[155,157],"using":[158],"Bayes'":[159],"rule":[160],"during":[161],"generation.":[162],"This":[163],"enables":[165,250],"probabilistic":[166],"re-ranking,":[167],"parallel":[168],"memory":[169],"usage,":[170],"clear":[172],"contribution,":[176],"well-suited":[179],"for":[180],"large":[181],"collections.":[183],"evaluate":[185],"BEFT":[188],"primarily":[189],"knowledge-based":[191],"tasks,":[195],"must":[198],"reason":[199],"over":[200,210],"long,":[201],"imperfect":[202],"retrieval":[203],"lists.":[204],"The":[205,234],"results":[206],"show":[207],"substantial":[208],"improvements":[209],"standard":[211,254],"RAG,":[212],"including":[213],"strong":[214],"gains":[215],"Document":[217],"Visual":[218],"Question":[219],"Answering":[220],"multimodal":[222],"needle-in-a-haystack":[223],"benchmarks.":[224],"demonstrate":[227],"mitigates":[230],"effect.":[233],"be":[238],"used":[239],"detect":[241],"insufficient":[242],"grounding":[243],"trigger":[245],"deflection,":[246],"while":[247],"pruning":[249],"faster":[251],"decoding":[252],"RAG.":[255]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-04-28T00:00:00"}