{"id":"https://openalex.org/W4382603217","doi":"https://doi.org/10.48550/arxiv.2306.15766","title":"Large Language Models as Annotators: Enhancing Generalization of NLP Models at Minimal Cost","display_name":"Large Language Models as Annotators: Enhancing Generalization of NLP Models at Minimal Cost","publication_year":2023,"publication_date":"2023-06-27","ids":{"openalex":"https://openalex.org/W4382603217","doi":"https://doi.org/10.48550/arxiv.2306.15766"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2306.15766","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.15766","pdf_url":"https://arxiv.org/pdf/2306.15766","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2306.15766","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Bansal, Parikshit","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210124949","display_name":"Microsoft Research (India)","ror":"https://ror.org/02w7f3w92","country_code":"IN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210124949"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Bansal, Parikshit","raw_affiliation_strings":["Microsoft Research India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research India","institution_ids":["https://openalex.org/I4210124949"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080592530","display_name":"Amit Kumar Sharma","orcid":"https://orcid.org/0000-0003-1451-5892","observed_orcids":["https://orcid.org/0000-0003-1451-5892"]},"institutions":[{"id":"https://openalex.org/I4210124949","display_name":"Microsoft Research (India)","ror":"https://ror.org/02w7f3w92","country_code":"IN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210124949"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sharma, Amit","raw_affiliation_strings":["Microsoft Research India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research India","institution_ids":["https://openalex.org/I4210124949"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210124949"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":7,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.7275000214576721,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.7275000214576721,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.15629999339580536,"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/T10028","display_name":"Topic Modeling","score":0.09839999675750732,"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/active-learning-machine-learning","display_name":"active learning (machine learning)","score":0.7429999709129333},{"id":"https://openalex.org/keywords/annotation-budget","display_name":"annotation budget","score":0.5879999995231628},{"id":"https://openalex.org/keywords/domain-generalization","display_name":"domain generalization","score":0.6420000195503235},{"id":"https://openalex.org/keywords/low-data-regimes","display_name":"low-data regimes","score":0.7120000123977661},{"id":"https://openalex.org/keywords/sampling-strategy","display_name":"sampling strategy","score":0.4350000023841858},{"id":"https://openalex.org/keywords/out-of-domain-generalization","display_name":"out-of-domain generalization","score":0.6340000033378601}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8705155849456787},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7572739124298096},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6928226351737976},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6050766110420227},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.583702802658081},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5647502541542053},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.551866888999939},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5371094346046448},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.5246270298957825},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.46863389015197754},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.4451368451118469},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.42756780982017517},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.41039231419563293},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2306.15766","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.15766","pdf_url":"https://arxiv.org/pdf/2306.15766","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2306.15766","doi":"https://doi.org/10.48550/arxiv.2306.15766","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2306.15766","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2306.15766","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.15766","pdf_url":"https://arxiv.org/pdf/2306.15766","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4382603217.pdf","grobid_xml":"https://content.openalex.org/works/W4382603217.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4390962271","https://openalex.org/W4416035993","https://openalex.org/W4388788754","https://openalex.org/W4389518684","https://openalex.org/W7125567162","https://openalex.org/W4417538289","https://openalex.org/W4389523900","https://openalex.org/W4390897950","https://openalex.org/W7133213760","https://openalex.org/W7155095126"],"abstract_inverted_index":{"State-of-the-art":[0],"supervised":[1],"NLP":[2,56,80,117,124],"models":[3,46,125],"achieve":[4],"high":[5],"accuracy":[6,152],"but":[7],"are":[8,22,126],"also":[9],"susceptible":[10],"to":[11,31,75,148],"failures":[12],"on":[13,104],"inputs":[14,50,74],"from":[15,128],"low-data":[16],"regimes,":[17],"such":[18,89],"as":[19,90],"domains":[20],"that":[21,84,122,143],"not":[23,94],"represented":[24],"in":[25,107,151],"training":[26,156],"data.":[27],"As":[28],"an":[29,67],"approximation":[30],"collecting":[32],"ground-truth":[33],"labels":[34],"for":[35,48,62,69,153],"the":[36,41,53,71,79,105,111,115,120,155],"specific":[37],"domain,":[38],"we":[39,65,98],"study":[40],"use":[42],"of":[43,55],"large":[44],"language":[45],"(LLMs)":[47],"annotating":[49],"and":[51,77,114,137,157],"improving":[52],"generalization":[54],"models.":[57],"Specifically,":[58],"given":[59],"a":[60,100,129],"budget":[61],"LLM":[63],"annotations,":[64],"present":[66],"algorithm":[68],"sampling":[70,92,101,145],"most":[72,123],"informative":[73],"annotate":[76],"retrain":[78],"model.":[81,131],"We":[82],"find":[83],"popular":[85],"active":[86],"learning":[87],"strategies":[88],"uncertainty-based":[91],"do":[93],"work":[95],"well.":[96],"Instead,":[97],"propose":[99],"strategy":[102,146],"based":[103],"difference":[106],"prediction":[108],"scores":[109],"between":[110],"base":[112,130],"model":[113],"finetuned":[116,127],"model,":[118],"utilizing":[119],"fact":[121],"Experiments":[132],"with":[133],"classification":[134],"(semantic":[135,139],"similarity)":[136],"ranking":[138],"search)":[140],"tasks":[141],"show":[142],"our":[144],"leads":[147],"significant":[149],"gains":[150],"both":[154],"target":[158],"domains.":[159]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":1}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2023-06-30T00:00:00"}