{"id":"https://openalex.org/W4416034351","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.827","title":"Token-Level Metrics for Detecting Incorrect Gold Annotations in Named Entity Recognition","display_name":"Token-Level Metrics for Detecting Incorrect Gold Annotations in Named Entity Recognition","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034351","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.827"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.827","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.827","pdf_url":"https://aclanthology.org/2025.findings-emnlp.827.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: EMNLP 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-emnlp.827.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5007507952","display_name":"Elena Merdjanovska","orcid":"https://orcid.org/0000-0003-0794-8438"},"institutions":[{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Elena Merdjanovska","raw_affiliation_strings":["Humboldt-Universitt zu Berlin Science of Intelligence Humboldt-Universitt zu Berlin Science of Intelligence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Humboldt-Universitt zu Berlin Science of Intelligence Humboldt-Universitt zu Berlin Science of Intelligence","institution_ids":["https://openalex.org/I39343248"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032877157","display_name":"Alan Akbik","orcid":null},"institutions":[{"id":"https://openalex.org/I39343248","display_name":"Humboldt-Universit\u00e4t zu Berlin","ror":"https://ror.org/01hcx6992","country_code":"DE","type":"education","lineage":["https://openalex.org/I39343248"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Alan Akbik","raw_affiliation_strings":["Humboldt-Universitt zu Berlin Science of Intelligence Humboldt-Universitt zu Berlin Science of Intelligence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Humboldt-Universitt zu Berlin Science of Intelligence Humboldt-Universitt zu Berlin Science of Intelligence","institution_ids":["https://openalex.org/I39343248"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I39343248"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.2982024,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15292","last_page":"15304"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5246999859809875,"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":0.5246999859809875,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.10679999738931656,"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/T11719","display_name":"Data Quality and Management","score":0.0869000032544136,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/named-entity-recognition","display_name":"Named-entity recognition","score":0.4715999960899353},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3050999939441681},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.3034000098705292},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.30149999260902405},{"id":"https://openalex.org/keywords/entity-linking","display_name":"Entity linking","score":0.2831999957561493}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6870999932289124},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5310999751091003},{"id":"https://openalex.org/C2779135771","wikidata":"https://www.wikidata.org/wiki/Q403574","display_name":"Named-entity recognition","level":3,"score":0.4715999960899353},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4323999881744385},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3050999939441681},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3034000098705292},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.30149999260902405},{"id":"https://openalex.org/C96711827","wikidata":"https://www.wikidata.org/wiki/Q17012245","display_name":"Entity linking","level":3,"score":0.2831999957561493},{"id":"https://openalex.org/C2777889803","wikidata":"https://www.wikidata.org/wiki/Q25047676","display_name":"Named entity","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.827","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.827","pdf_url":"https://aclanthology.org/2025.findings-emnlp.827.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.827","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.827","pdf_url":"https://aclanthology.org/2025.findings-emnlp.827.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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5729854464","display_name":"Eidetische Repr\u00e4sentationen Nat\u00fcrlicher Sprache","funder_award_id":"448414230","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G6510858048","display_name":"EXC 2002: Science of Intelligence (SCIoI)","funder_award_id":"390523135","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G6532450817","display_name":null,"funder_award_id":"EXC 2002","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"}],"funders":[{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034351.pdf","grobid_xml":"https://content.openalex.org/works/W4416034351.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Annotated":[0],"datasets":[1],"for":[2,29,70,90],"supervised":[3],"learning":[4,20],"tasks":[5],"often":[6],"contain":[7],"incorrect":[8,72,96,115],"gold":[9],"annotations,":[10],"i.e.":[11],"label":[12,19,62,144],"noise.To":[13],"address":[14],"this":[15],"issue,":[16],"many":[17],"noisy":[18],"approaches":[21],"incorporate":[22],"metrics":[23,42,69,87,102,106,134,140],"to":[24,51,60,161],"filter":[25],"out":[26],"unreliable":[27],"samples,":[28],"example":[30],"using":[31,126],"heuristics":[32],"such":[33],"as":[34,82],"high":[35],"loss":[36],"or":[37],"low":[38],"confidence.However,":[39],"when":[40,122],"these":[41,165],"are":[43,88,98,158],"integrated":[44],"into":[45],"larger":[46],"pipelines,":[47],"it":[48],"becomes":[49],"difficult":[50],"compare":[52,104],"their":[53,57,110],"effectiveness,":[54],"and":[55,94,117,163],"understand":[56],"individual":[58],"contribution":[59],"reducing":[61],"noise.This":[63],"paper":[64],"directly":[65],"compares":[66],"popular":[67],"sample":[68],"detecting":[71,113],"annotations":[73],"in":[74,112],"named":[75],"entity":[76],"recognition":[77],"(NER).NER":[78],"is":[79,167],"commonly":[80],"approached":[81],"token":[83,93],"classification,":[84],"so":[85],"the":[86,95,105,114,119,127,136,143,152],"calculated":[89],"each":[91],"training":[92,132],"ones":[97],"flagged":[99],"by":[100],"defining":[101],"thresholds.We":[103],"based":[107],"on":[108],"(i)":[109],"accuracy":[111],"labels":[116],"(ii)":[118],"test":[120],"scores":[121],"retraining":[123],"a":[124,168],"model":[125,153],"cleaned":[128],"dataset.We":[129],"show":[130],"that":[131,151],"dynamics":[133],"work":[135],"best":[137,139],"overall.The":[138],"effectively":[141],"reduce":[142],"noise":[145,148],"across":[146],"different":[147],"types.The":[149],"errors":[150],"has":[154],"not":[155],"yet":[156],"memorized":[157],"more":[159,169],"feasible":[160],"detect,":[162],"relabeling":[164],"tokens":[166],"effective":[170],"strategy":[171],"than":[172],"excluding":[173],"them":[174],"from":[175],"training.":[176],"1":[177]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-08T00:00:00"}