{"id":"https://openalex.org/W3191689059","doi":null,"title":"Mitigating Bias in the SNLI Dataset","display_name":"Mitigating Bias in the SNLI Dataset","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W3191689059","mag":"3191689059"},"language":"en","primary_location":{"id":"pmh:oai:dspace.library.uu.nl:1874/404102","doi":null,"is_oa":true,"landing_page_url":"http://dspace.library.uu.nl/handle/1874/404102","pdf_url":"http://dspace.library.uu.nl/handle/1874/404102","source":{"id":"https://openalex.org/S4306401649","display_name":"Utrecht University Repository (Utrecht University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I193662353","host_organization_name":"Utrecht University","host_organization_lineage":["https://openalex.org/I193662353"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Bachelor thesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://dspace.library.uu.nl/handle/1874/404102","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5036434528","display_name":"Albert Baas","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Albert Baas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5036434528"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":0.7465999722480774,"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":0.7465999722480774,"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":0.15800000727176666,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.03440000116825104,"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/natural-language-inference","display_name":"natural language inference","score":0.8199999928474426},{"id":"https://openalex.org/keywords/annotation-artifacts","display_name":"annotation artifacts","score":0.9380000233650208},{"id":"https://openalex.org/keywords/bias-mitigation","display_name":"bias mitigation","score":0.8100000023841858},{"id":"https://openalex.org/keywords/data-augmentation","display_name":"data augmentation","score":0.9470000267028809},{"id":"https://openalex.org/keywords/dataset-bias","display_name":"data bias","score":0.375},{"id":"https://openalex.org/keywords/logical-consequence","display_name":"logical consequence","score":0.6420000195503235},{"id":"https://openalex.org/keywords/contradiction","display_name":"contradiction","score":0.6959999799728394},{"id":"https://openalex.org/keywords/neutrals","display_name":"neutrals","score":0.5370000004768372},{"id":"https://openalex.org/keywords/distilbert","display_name":"DistilBERT","score":0.9359999895095825}],"concepts":[{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.4272807240486145},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3209998607635498}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:dspace.library.uu.nl:1874/404102","doi":null,"is_oa":true,"landing_page_url":"http://dspace.library.uu.nl/handle/1874/404102","pdf_url":"http://dspace.library.uu.nl/handle/1874/404102","source":{"id":"https://openalex.org/S4306401649","display_name":"Utrecht University Repository (Utrecht University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I193662353","host_organization_name":"Utrecht University","host_organization_lineage":["https://openalex.org/I193662353"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Bachelor thesis"},{"id":"mag:3191689059","doi":null,"is_oa":false,"landing_page_url":"https://dspace.library.uu.nl/handle/1874/404102","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:dspace.library.uu.nl:1874/404102","doi":null,"is_oa":true,"landing_page_url":"http://dspace.library.uu.nl/handle/1874/404102","pdf_url":"http://dspace.library.uu.nl/handle/1874/404102","source":{"id":"https://openalex.org/S4306401649","display_name":"Utrecht University Repository (Utrecht University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I193662353","host_organization_name":"Utrecht University","host_organization_lineage":["https://openalex.org/I193662353"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Bachelor thesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3191689059.pdf","grobid_xml":"https://content.openalex.org/works/W3191689059.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W1840435438","https://openalex.org/W2081580037","https://openalex.org/W2185175083","https://openalex.org/W2753316031","https://openalex.org/W2790415926","https://openalex.org/W2798358706","https://openalex.org/W2896317853","https://openalex.org/W2955503152","https://openalex.org/W2971296908","https://openalex.org/W2978017171","https://openalex.org/W3023533951","https://openalex.org/W3034584102","https://openalex.org/W3034850762","https://openalex.org/W3132011002"],"related_works":["https://openalex.org/W3009380369","https://openalex.org/W2952341934","https://openalex.org/W2984256198","https://openalex.org/W2955503152","https://openalex.org/W3092298059","https://openalex.org/W4405765350","https://openalex.org/W7119470366","https://openalex.org/W3197317199","https://openalex.org/W3016924538","https://openalex.org/W3034584102"],"abstract_inverted_index":{"Natural":[0,59],"language":[1],"inference":[2,23],"(NLI)":[3],"is":[4,12,31,199],"the":[5,28,33,36,38,50,57,72,78,91,113,127,131,150,160,165,169,180,193],"task":[6],"of":[7,52,71,179],"determining":[8],"whether":[9],"a":[10,21,106,147,209],"\"hypothesis\"":[11],"true":[13],"(entailment),":[14],"false":[15],"(contradiction),":[16],"or":[17],"undetermined":[18],"(neutral)":[19],"given":[20],"\"premise\".An":[22],"contains":[24],"two":[25],"sentences,":[26],"where":[27],"first":[29],"sentence":[30],"considered":[32],"premise":[34],"and":[35,175],"second":[37],"hypothesis.NLI":[39],"models":[40,54,83],"have":[41,48],"been":[42],"performing":[43],"very":[44],"well.However,":[45],"recent":[46],"studies":[47],"shown":[49],"performance":[51,186],"NLI":[53,82,182],"trained":[55],"on":[56,86,159,189],"Stanford":[58],"Language":[60],"Inference":[61],"(SNLI)":[62],"dataset":[63,172],"are":[64,101],"overestimated.Hypothesis-only":[65],"classifiers":[66],"can":[67],"already":[68],"label":[69],"most":[70],"inferences":[73],"correctly":[74],"without":[75],"looking":[76],"at":[77],"premise.This":[79],"suggests":[80],"that":[81,149,164,176],"depend":[84],"heavily":[85],"annotation":[87,99,128,194],"artifacts":[88,100,129,195],"present":[89],"in":[90,103,146,168,208],"hypothesis.A":[92],"bias":[93,167,207],"seems":[94],"to":[95,105,125,142,154,204],"occur":[96],"because":[97],"certain":[98],"often":[102],"relation":[104],"specific":[107],"label.This":[108],"thesis":[109],"will":[110,122,139,152],"explore":[111],"debiasing":[112],"SNLI":[114,171],"dataset,":[115],"which":[116],"has":[117,173],"this":[118,190],"hypothesis-only":[119,166,206],"bias.New":[120],"examples":[121],"be":[123,140],"generated":[124],"balance":[126],"over":[130,196],"different":[132,156],"labels.A":[133],"few":[134],"simple":[135],"data":[136],"augmentation":[137],"techniques":[138],"used":[141],"generate":[143],"new":[144],"premises":[145],"way":[148,203],"hypotheses":[151],"relate":[153],"3":[155],"labels,":[157],"depending":[158],"premise.The":[161],"results":[162],"show":[163],"augmented":[170,191],"decreased":[174],"DistilBERT,":[177],"one":[178],"high-performing":[181],"models,":[183],"maintains":[184],"similar":[185],"after":[187],"fine-tuning":[188],"dataset.Balancing":[192],"all":[197],"labels":[198],"therefore":[200],"an":[201],"efficient":[202],"mitigate":[205],"dataset.":[210]},"counts_by_year":[],"updated_date":"2026-09-26T08:19:03.415552","created_date":"2021-08-16T00:00:00"}