{"id":"https://openalex.org/W4416694370","doi":"https://doi.org/10.48550/arxiv.2509.01564","title":"Enhancing Uncertainty Estimation in LLMs with Expectation of Aggregated Internal Belief","display_name":"Enhancing Uncertainty Estimation in LLMs with Expectation of Aggregated Internal Belief","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4416694370","doi":"https://doi.org/10.48550/arxiv.2509.01564"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2509.01564","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2509.01564","pdf_url":"https://arxiv.org/pdf/2509.01564","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2509.01564","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090522275","display_name":"Zeguan Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Zeguan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Dou, Diyang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dou, Diyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101260675","display_name":"Boya Xiong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiong, Boya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100679129","display_name":"Yun Chen","orcid":"https://orcid.org/0009-0006-2247-6458"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5114269723","display_name":"Guanhua Chen","orcid":"https://orcid.org/0009-0009-1764-8384"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Guanhua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"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/T10028","display_name":"Topic Modeling","score":0.3248000144958496,"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":0.3248000144958496,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.1597999930381775,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.10019999742507935,"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/overconfidence-effect","display_name":"Overconfidence effect","score":0.9544000029563904},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.671999990940094},{"id":"https://openalex.org/keywords/eagle","display_name":"Eagle","score":0.5407000184059143},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.5091000199317932},{"id":"https://openalex.org/keywords/confidence-interval","display_name":"Confidence interval","score":0.454800009727478},{"id":"https://openalex.org/keywords/estimation","display_name":"Estimation","score":0.41269999742507935}],"concepts":[{"id":"https://openalex.org/C51110983","wikidata":"https://www.wikidata.org/wiki/Q16503490","display_name":"Overconfidence effect","level":2,"score":0.9544000029563904},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.671999990940094},{"id":"https://openalex.org/C2780985876","wikidata":"https://www.wikidata.org/wiki/Q2092297","display_name":"Eagle","level":2,"score":0.5407000184059143},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.5091000199317932},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5049999952316284},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5023999810218811},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.4925999939441681},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4729999899864197},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.454800009727478},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4056999981403351},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.3538999855518341},{"id":"https://openalex.org/C28427503","wikidata":"https://www.wikidata.org/wiki/Q13580300","display_name":"Internal model","level":3,"score":0.3443000018596649},{"id":"https://openalex.org/C2776035688","wikidata":"https://www.wikidata.org/wiki/Q1606558","display_name":"Affect (linguistics)","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C2909755999","wikidata":"https://www.wikidata.org/wiki/Q4751126","display_name":"Low Confidence","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C12174686","wikidata":"https://www.wikidata.org/wiki/Q1058438","display_name":"Risk assessment","level":2,"score":0.26980000734329224},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2551000118255615}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2509.01564","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2509.01564","pdf_url":"https://arxiv.org/pdf/2509.01564","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"pmh:oai:ojs.aaai.org:article/40698","is_oa":false,"landing_page_url":"https://ojs.aaai.org/index.php/AAAI/article/view/40698","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2159-5399","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"doi:10.48550/arxiv.2509.01564","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2509.01564","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":"pmh:oai:arXiv.org:2509.01564","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2509.01564","pdf_url":"https://arxiv.org/pdf/2509.01564","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320309612","display_name":"Natural Science Foundation of Shanghai","ror":null},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320327605","display_name":"Southern University of Science and Technology","ror":null},{"id":"https://openalex.org/F4320337111","display_name":"Basic and Applied Basic Research Foundation of Guangdong Province","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"achieved":[5],"remarkable":[6],"success":[7],"across":[8],"a":[9,59,114,152,158],"wide":[10],"range":[11],"of":[12,55,70,79,149,155,160,163,169,172],"natural":[13],"language":[14],"tasks,":[15],"but":[16],"often":[17],"exhibit":[18],"overconfidence":[19],"and":[20,45,102,131,166],"generate":[21],"plausible":[22],"yet":[23],"incorrect":[24],"answers.":[25],"This":[26],"overconfidence,":[27],"especially":[28],"in":[29],"models":[30],"undergone":[31],"Reinforcement":[32],"Learning":[33],"from":[34,91],"Human":[35],"Feedback":[36],"(RLHF),":[37],"poses":[38],"significant":[39],"challenges":[40],"for":[41],"reliable":[42],"uncertainty":[43,156],"estimation":[44],"safe":[46],"deployment.":[47],"In":[48],"this":[49],"paper,":[50],"we":[51],"propose":[52],"EAGLE":[53,112,135],"(Expectation":[54],"AGgregated":[56],"internaL":[57],"bEief),":[58],"novel":[60],"self-evaluation-based":[61],"calibration":[62,138],"method":[63],"that":[64,118,134],"leverages":[65],"the":[66,82,104,107,122,161,170],"internal":[67,89,124],"hidden":[68],"states":[69],"LLMs":[71,132],"to":[72],"derive":[73],"more":[74,119],"accurate":[75],"confidence":[76,109,116],"scores.":[77],"Instead":[78],"relying":[80],"on":[81,128],"model's":[83,123],"final":[84],"output,":[85],"our":[86],"approach":[87],"extracts":[88],"beliefs":[90,101],"multiple":[92],"intermediate":[93],"layers":[94],"during":[95],"self-evaluation.":[96],"By":[97],"aggregating":[98],"these":[99],"layer-wise":[100,153],"calculating":[103],"expectation":[105],"over":[106,140],"resulting":[108],"score":[110,117,174],"distribution,":[111],"produces":[113],"refined":[115],"faithfully":[120],"reflects":[121],"certainty.":[125],"Extensive":[126],"experiments":[127],"diverse":[129],"datasets":[130],"demonstrate":[133],"significantly":[136],"improves":[137],"performance":[139],"existing":[141],"baselines.":[142],"We":[143],"also":[144],"provide":[145],"an":[146,167],"in-depth":[147],"analysis":[148,168],"EAGLE,":[150],"including":[151],"examination":[154],"patterns,":[157],"study":[159],"impact":[162],"self-evaluation":[164,173],"prompts,":[165],"effect":[171],"range.":[175]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}