{"id":"https://openalex.org/W4412888333","doi":"https://doi.org/10.18653/v1/2025.findings-acl.603","title":"Large Language Models are Miscalibrated In-Context Learners","display_name":"Large Language Models are Miscalibrated In-Context Learners","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888333","doi":"https://doi.org/10.18653/v1/2025.findings-acl.603"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.603","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.603","pdf_url":"https://aclanthology.org/2025.findings-acl.603.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.603.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5070178016","display_name":"Chengzu Li","orcid":"https://orcid.org/0000-0002-1879-7391"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chengzu Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111207827","display_name":"Han Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han Zhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079336821","display_name":"Goran Glava\u009a\u0161","orcid":"https://orcid.org/0000-0002-1301-6314"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Goran Glava\u0161","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081393566","display_name":"Anna Korhonen","orcid":"https://orcid.org/0000-0002-3692-3144"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anna Korhonen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5014866912","display_name":"Ivan Vuli\u0107","orcid":"https://orcid.org/0000-0002-5161-5422"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ivan Vuli\u0107","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":1.673,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.83379731,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"11575","last_page":"11596"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9567000269889832,"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":0.9567000269889832,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9455000162124634,"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/computer-science","display_name":"Computer science","score":0.7364398241043091},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5919291973114014},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.48972395062446594},{"id":"https://openalex.org/keywords/context-model","display_name":"Context model","score":0.44581979513168335},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.4073154330253601},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3526081442832947},{"id":"https://openalex.org/keywords/history","display_name":"History","score":0.10929885506629944}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7364398241043091},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5919291973114014},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.48972395062446594},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.44581979513168335},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.4073154330253601},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3526081442832947},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.10929885506629944},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.603","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.603","pdf_url":"https://aclanthology.org/2025.findings-acl.603.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.603","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.603","pdf_url":"https://aclanthology.org/2025.findings-acl.603.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7699999809265137,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888333.pdf","grobid_xml":"https://content.openalex.org/works/W4412888333.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W3204019825","https://openalex.org/W2349222429","https://openalex.org/W2116230991","https://openalex.org/W3117430770","https://openalex.org/W2046058552","https://openalex.org/W2590751808","https://openalex.org/W1972377868","https://openalex.org/W2132709506","https://openalex.org/W2098730537","https://openalex.org/W2186895195"],"abstract_inverted_index":{"When":[0],"adapting":[1],"ICL":[2,150],"with":[3,134],"or":[4,110,114,127,214],"without":[5,22],"fine-tuning,":[6],"we":[7,45,70,93,204],"are":[8],"curious":[9],"about":[10],"whether":[11,176],"the":[12,25,51,60,73,96,120,144,152,177,183,210],"instructiontuned":[13],"language":[14],"model":[15,184],"is":[16],"able":[17],"to":[18,118,164,194],"achieve":[19,84],"wellcalibrated":[20],"results":[21],"suffering":[23],"from":[24,59],"problem":[26,75,147],"of":[27,50,56,62,98,107,148,201],"overconfidence":[28],"(i.e.,":[29],"miscalibration)":[30],"considering":[31],"its":[32],"strong":[33],"instruction":[34],"following":[35],"ability,":[36],"especially":[37],"in":[38,81,112,154],"such":[39],"limited":[40],"data":[41,178],"setups.In":[42],"this":[43],"work,":[44],"deliver":[46],"an":[47],"in-depth":[48],"analysis":[49],"behavior":[52],"across":[53,77],"different":[54,102,115],"choices":[55],"learning":[57,79,162,172],"methods":[58,80],"perspective":[61],"both":[63,88,196],"performance":[64,90,156,198],"and":[65,91,124,139,157,186,199],"calibration.Through":[66],"extensive":[67],"controlled":[68],"experiments,":[69],"observe":[71],"that":[72,132],"miscalibration":[74],"exists":[76],"all":[78],"low-resource":[82],"setups.To":[83],"simultaneous":[85],"gain":[86],"for":[87,170],"in-task":[89],"calibration,":[92],"then":[94],"study":[95],"potential":[97,145],"self-ensembling":[99,133,191],"applied":[100],"at":[101],"modeling":[103],"stages":[104],"(e.g.,":[105],"variations":[106,111],"incontext":[108],"examples":[109],"prompts":[113],"ensembling":[116],"strategies)":[117],"make":[119],"predictions":[121],"more":[122],"calibrated":[123,140],"have":[125],"comparable":[126],"even":[128],"better":[129],"performance.We":[130],"find":[131],"max":[135],"probability":[136],"produces":[137],"robust":[138],"predictions.Our":[141],"work":[142],"reveals":[143],"calibration":[146,200],"using":[149],"despite":[151],"improvements":[153],"task":[155,197],"sheds":[158],"light":[159],"on":[160,175,192],"which":[161,203],"paradigm":[163],"choose.We":[165],"also":[166],"provide":[167],"practical":[168],"guidelines":[169],"choosing":[171],"paradigms":[173],"depending":[174],"has":[179],"been":[180],"seen":[181],"by":[182],"before":[185],"a":[187],"worthwhile":[188],"solution":[189],"via":[190],"how":[193],"enhance":[195],"LMs,":[202],"hope":[205],"could":[206],"encourage":[207],"further":[208],"study.Is":[209],"sentence":[211],"hate,":[212],"neutral":[213],"support?<SENTENCE>":[215]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}