{"id":"https://openalex.org/W7167742306","doi":"https://doi.org/10.48550/arxiv.2607.05721","title":"SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation","display_name":"SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation","publication_year":2026,"publication_date":"2026-07-07","ids":{"openalex":"https://openalex.org/W7167742306","doi":"https://doi.org/10.48550/arxiv.2607.05721"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.05721","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.05721","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":"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.2607.05721","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140234766","display_name":"Yimeng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yimeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140283922","display_name":"Yingying Zhuang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuang, Yingying","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140274099","display_name":"Ziyi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Ziyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085698173","display_name":"Ye Lu","orcid":"https://orcid.org/0000-0002-2376-4519"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Yuxuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140280239","display_name":"Pei Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Pei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061662922","display_name":"A. Gupta","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gupta, Aman","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101440908","display_name":"SU Zhe","orcid":"https://orcid.org/0009-0002-2661-7722"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Su, Zhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127944120","display_name":"Ming Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Ming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140246923","display_name":"Zhilin Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhilin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140267818","display_name":"Qun Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Qun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120309268","display_name":"Manikandarajan Ramanathan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ramanathan, Manikandarajan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119875936","display_name":"Rajashekar Maragoud","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maragoud, Rajashekar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090905459","display_name":"Edward Vul","orcid":"https://orcid.org/0000-0002-6217-1273"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vul, Edward","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140232099","display_name":"Jing Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Jing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5140290413","display_name":"Dakuo Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Dakuo","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":0.43470001220703125,"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":0.43470001220703125,"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":0.10429999977350235,"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":0.07150000333786011,"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/benchmark","display_name":"Benchmark (surveying)","score":0.582099974155426},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.5393999814987183},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5375000238418579},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.492900013923645},{"id":"https://openalex.org/keywords/uncertainty-quantification","display_name":"Uncertainty quantification","score":0.49140000343322754},{"id":"https://openalex.org/keywords/heuristic","display_name":"Heuristic","score":0.46619999408721924},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.46380001306533813},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.41929998993873596},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3815999925136566},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.364300012588501}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7124999761581421},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.583899974822998},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.582099974155426},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.5393999814987183},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5375000238418579},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.492900013923645},{"id":"https://openalex.org/C32230216","wikidata":"https://www.wikidata.org/wiki/Q7882499","display_name":"Uncertainty quantification","level":2,"score":0.49140000343322754},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.46619999408721924},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4643000066280365},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.46380001306533813},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.4593999981880188},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.41929998993873596},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3815999925136566},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.364300012588501},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3580000102519989},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.3531000018119812},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.3434999883174896},{"id":"https://openalex.org/C137209882","wikidata":"https://www.wikidata.org/wiki/Q1403517","display_name":"Measurement uncertainty","level":2,"score":0.3328999876976013},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.3327000141143799},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3066999912261963},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.2969000041484833},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.29120001196861267},{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.28850001096725464},{"id":"https://openalex.org/C917703","wikidata":"https://www.wikidata.org/wiki/Q7239668","display_name":"Predictive inference","level":5,"score":0.2851000130176544},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.2705000042915344},{"id":"https://openalex.org/C153701036","wikidata":"https://www.wikidata.org/wiki/Q659974","display_name":"Trustworthiness","level":2,"score":0.26840001344680786},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.26579999923706055},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.2572000026702881},{"id":"https://openalex.org/C199465337","wikidata":"https://www.wikidata.org/wiki/Q6823168","display_name":"Metaproteomics","level":4,"score":0.25679999589920044},{"id":"https://openalex.org/C27158222","wikidata":"https://www.wikidata.org/wiki/Q5532422","display_name":"Generalizability theory","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.05721","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.05721","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":"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.2607.05721","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.05721","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":"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":{"Uncertainty":[0,47],"estimation":[1],"is":[2],"essential":[3],"not":[4],"only":[5],"for":[6,20,58],"the":[7,55,84,136,166,172,192],"trustworthy":[8],"deployment":[9],"of":[10,70,117,125],"large":[11],"language":[12],"models":[13],"(LLMs)":[14],"but":[15],"also":[16],"as":[17],"a":[18,50,66,79,92,102,115,122],"foundation":[19],"self-refinement":[21],"in":[22],"LLM":[23,97,159],"generation.":[24],"However,":[25],"existing":[26],"approaches":[27],"operate":[28],"at":[29],"suboptimal":[30],"granularities:":[31],"token-level":[32],"scores":[33,39],"lack":[34],"semantic":[35],"coherence,":[36],"while":[37,180],"sequence-level":[38,202],"fail":[40],"to":[41,106],"localize":[42],"errors.":[43],"We":[44,133],"formalize":[45],"Span-Level":[46],"Estimation":[48],"(SLUE),":[49],"new":[51],"task":[52],"that":[53,82,162,201],"targets":[54],"natural":[56],"granularity":[57],"uncertainty:":[59],"semantically":[60],"coherent":[61],"text":[62],"spans,":[63,146],"each":[64],"conveying":[65],"single":[67,93],"assessable":[68],"unit":[69],"meaning.":[71],"To":[72],"address":[73],"this":[74],"task,":[75],"we":[76],"introduce":[77],"SPANUQ,":[78],"lightweight":[80],"probe":[81,174],"distills":[83],"uncertainty":[85,113,139,169],"knowledge":[86],"from":[87,152],"expensive":[88],"multi-sample":[89,153],"inference":[90],"into":[91],"forward":[94],"pass":[95],"over":[96],"hidden":[98],"states.":[99],"SPANUQ":[100,163],"employs":[101],"DETR-style":[103],"span":[104,186],"decoder":[105],"simultaneously":[107],"detect":[108],"spans":[109],"and":[110,129,147,176],"estimate":[111],"their":[112],"via":[114],"Mixture":[116],"Beta":[118,126],"distribution,":[119],"trained":[120],"with":[121],"principled":[123],"combination":[124],"NLL":[127],"regression":[128],"contrastive":[130],"ranking":[131],"objectives.":[132],"construct":[134],"SPANUQ-BENCH,":[135],"first":[137],"span-level":[138,168],"benchmark":[140],"comprising":[141],"20K":[142],"prompts,":[143],"293K":[144],"annotated":[145],"continuous":[148],"soft":[149],"labels":[150],"derived":[151],"claim":[154],"verification.":[155],"Experiments":[156],"on":[157],"five":[158,210],"backbones":[160],"show":[161],"consistently":[164],"achieves":[165],"best":[167,193],"quality,":[170],"outperforming":[171],"strongest":[173],"baseline":[175],"all":[177],"sampling-based":[178],"methods":[179,203],"being":[181],"10-20x":[182],"faster.":[183],"Its":[184],"DETR-based":[185],"detector":[187],"attains":[188],"0.910":[189],"F1,":[190],"surpassing":[191],"heuristic":[194],"by":[195],"39.4%,":[196],"enabling":[197],"precise":[198],"error":[199],"localization":[200],"cannot":[204],"provide.":[205],"The":[206],"framework":[207],"generalizes":[208],"across":[209],"LLMs":[211],"spanning":[212],"two":[213],"model":[214],"families.":[215]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-09T00:00:00"}