{"id":"https://openalex.org/W7155017826","doi":"https://doi.org/10.48550/arxiv.2604.15741","title":"Learning Uncertainty from Sequential Internal Dispersion in Large Language Models","display_name":"Learning Uncertainty from Sequential Internal Dispersion in Large Language Models","publication_year":2026,"publication_date":"2026-04-17","ids":{"openalex":"https://openalex.org/W7155017826","doi":"https://doi.org/10.48550/arxiv.2604.15741"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.15741","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.15741","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,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.15741","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5106011068","display_name":"Ponhvoan Srey","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Srey, Ponhvoan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134056003","display_name":"Xiaobao Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xiaobao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134056082","display_name":"Cong-Duy Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nguyen, Cong-Duy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134012891","display_name":"Anh Tuan Luu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luu, Anh Tuan","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/T10028","display_name":"Topic Modeling","score":0.2587999999523163,"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":0.2587999999523163,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.06560000032186508,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.06440000236034393,"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/variance","display_name":"Variance (accounting)","score":0.7296000123023987},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.6200000047683716},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5576000213623047},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.5216000080108643},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4925999939441681},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.43560001254081726},{"id":"https://openalex.org/keywords/dispersion","display_name":"Dispersion (optics)","score":0.42160001397132874},{"id":"https://openalex.org/keywords/source-code","display_name":"Source code","score":0.34529998898506165}],"concepts":[{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.7296000123023987},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6837999820709229},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.6200000047683716},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5726000070571899},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5576000213623047},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.5216000080108643},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4925999939441681},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4860999882221222},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.43560001254081726},{"id":"https://openalex.org/C177562468","wikidata":"https://www.wikidata.org/wiki/Q182893","display_name":"Dispersion (optics)","level":2,"score":0.42160001397132874},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3549000024795532},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.34529998898506165},{"id":"https://openalex.org/C40506919","wikidata":"https://www.wikidata.org/wiki/Q7452469","display_name":"Sequence learning","level":2,"score":0.3327000141143799},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.30379998683929443},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.30160000920295715},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.296099990606308},{"id":"https://openalex.org/C28427503","wikidata":"https://www.wikidata.org/wiki/Q13580300","display_name":"Internal model","level":3,"score":0.29319998621940613},{"id":"https://openalex.org/C2988416141","wikidata":"https://www.wikidata.org/wiki/Q6031139","display_name":"Information loss","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27230000495910645},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2709999978542328},{"id":"https://openalex.org/C3020402766","wikidata":"https://www.wikidata.org/wiki/Q104376712","display_name":"Prior information","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.25440001487731934}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.15741","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.15741","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.15741","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.15741","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,"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"},"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":{"Uncertainty":[0],"estimation":[1],"is":[2,163],"a":[3,62,78],"promising":[4],"approach":[5],"to":[6,22],"detect":[7],"hallucinations":[8],"in":[9,85],"large":[10,151],"language":[11],"models":[12],"(LLMs).":[13],"Recent":[14],"approaches":[15],"commonly":[16],"depend":[17],"on":[18,31,46,100,150],"model":[19,107],"internal":[20,93],"states":[21,34],"estimate":[23],"uncertainty.":[24],"However,":[25],"they":[26],"suffer":[27],"from":[28,40,73],"strict":[29],"assumptions":[30],"how":[32],"hidden":[33,74],"should":[35],"evolve":[36],"across":[37,95],"layers,":[38,96],"and":[39,108,128,147],"information":[41,131],"loss":[42],"by":[43],"solely":[44],"focusing":[45],"last":[47],"or":[48,90],"mean":[49],"tokens.":[50],"To":[51],"address":[52],"these":[53],"issues,":[54],"we":[55],"present":[56],"Sequential":[57],"Internal":[58],"Variance":[59],"Representation":[60],"(SIVR),":[61],"supervised":[63],"hallucination":[64],"detection":[65],"framework":[66],"that":[67,82],"leverages":[68],"token-wise,":[69],"layer-wise":[70],"features":[71],"derived":[72],"states.":[75],"SIVR":[76,136,143],"adopts":[77],"more":[79],"basic":[80],"assumption":[81],"uncertainty":[83],"manifests":[84],"the":[86,105,114,155],"degree":[87],"of":[88,92,117,125],"dispersion":[89],"variance":[91,119],"representations":[94],"rather":[97],"than":[98],"relying":[99,149],"specific":[101],"assumptions,":[102],"which":[103],"makes":[104],"method":[106],"task":[109],"agnostic.":[110],"It":[111],"additionally":[112],"aggregates":[113],"full":[115],"sequence":[116],"per-token":[118],"features,":[120],"learning":[121],"temporal":[122],"patterns":[123],"indicative":[124],"factual":[126],"errors":[127],"thereby":[129],"preventing":[130],"loss.":[132],"Experimental":[133],"results":[134],"demonstrate":[135],"consistently":[137],"outperforms":[138],"strong":[139],"baselines.":[140],"Most":[141],"importantly,":[142],"enjoys":[144],"stronger":[145],"generalisation":[146],"avoids":[148],"training":[152],"sets,":[153],"highlighting":[154],"potential":[156],"for":[157],"practical":[158],"deployment.":[159],"Our":[160],"code":[161],"repository":[162],"available":[164],"online":[165],"at":[166],"https://github.com/ponhvoan/internal-variance.":[167]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-21T00:00:00"}