{"id":"https://openalex.org/W4385571347","doi":"https://doi.org/10.18653/v1/2023.acl-long.801","title":"Improving the robustness of NLI models with minimax training","display_name":"Improving the robustness of NLI models with minimax training","publication_year":2023,"publication_date":"2023-01-01","ids":{"openalex":"https://openalex.org/W4385571347","doi":"https://doi.org/10.18653/v1/2023.acl-long.801"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2023.acl-long.801","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.acl-long.801","pdf_url":"https://aclanthology.org/2023.acl-long.801.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":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2023.acl-long.801.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063100771","display_name":"Michalis Korakakis","orcid":null},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Michalis Korakakis","raw_affiliation_strings":["University of Cambridge"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5067943980","display_name":"Andreas Vlachos","orcid":"https://orcid.org/0000-0003-2123-5071"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Andreas Vlachos","raw_affiliation_strings":["University of Cambridge"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Cambridge","institution_ids":["https://openalex.org/I241749"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5063100771"],"corresponding_institution_ids":["https://openalex.org/I241749"],"apc_list":null,"apc_paid":null,"fwci":0.6797,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.75353985,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"14322","last_page":"14339"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9904000163078308,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9904000163078308,"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":0.9797000288963318,"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":0.923799991607666,"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":0.8301340341567993},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.7593937516212463},{"id":"https://openalex.org/keywords/minimax","display_name":"Minimax","score":0.7150924205780029},{"id":"https://openalex.org/keywords/weighting","display_name":"Weighting","score":0.6289267539978027},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5996183753013611},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.51543128490448},{"id":"https://openalex.org/keywords/adversarial-system","display_name":"Adversarial system","score":0.5103816390037537},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.46387919783592224},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.4371103048324585},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.322904109954834},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.23455405235290527},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11569583415985107}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8301340341567993},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7593937516212463},{"id":"https://openalex.org/C149728462","wikidata":"https://www.wikidata.org/wiki/Q751319","display_name":"Minimax","level":2,"score":0.7150924205780029},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.6289267539978027},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5996183753013611},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.51543128490448},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.5103816390037537},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.46387919783592224},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4371103048324585},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.322904109954834},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.23455405235290527},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11569583415985107},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.0},{"id":"https://openalex.org/C126838900","wikidata":"https://www.wikidata.org/wiki/Q77604","display_name":"Radiology","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2023.acl-long.801","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.acl-long.801","pdf_url":"https://aclanthology.org/2023.acl-long.801.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":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:www.repository.cam.ac.uk:1810/361115","is_oa":true,"landing_page_url":"https://www.repository.cam.ac.uk/handle/1810/361115","pdf_url":"https://www.repository.cam.ac.uk/bitstreams/dff96b3c-2359-420c-9dc4-fca073e68a94/download","source":{"id":"https://openalex.org/S4306401777","display_name":"Apollo (University of Cambridge)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I241749","host_organization_name":"University of Cambridge","host_organization_lineage":["https://openalex.org/I241749"],"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":"Conference Object"},{"id":"doi:10.17863/cam.104186","is_oa":true,"landing_page_url":"https://doi.org/10.17863/cam.104186","pdf_url":null,"source":{"id":"https://openalex.org/S7407050737","display_name":"Apollo","issn_l":null,"issn":[],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article"}],"best_oa_location":{"id":"doi:10.18653/v1/2023.acl-long.801","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.acl-long.801","pdf_url":"https://aclanthology.org/2023.acl-long.801.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":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.6499999761581421,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321848","display_name":"Cambridge Trust","ror":"https://ror.org/05sprqy15"},{"id":"https://openalex.org/F4320334630","display_name":"Economic and Social Research Council","ror":"https://ror.org/03n0ht308"}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4385571347.pdf"},"referenced_works_count":72,"referenced_works":["https://openalex.org/W1484551447","https://openalex.org/W1522301498","https://openalex.org/W1752492850","https://openalex.org/W1840435438","https://openalex.org/W2116064496","https://openalex.org/W2130158090","https://openalex.org/W2132984949","https://openalex.org/W2180093461","https://openalex.org/W2250539671","https://openalex.org/W2251386628","https://openalex.org/W2515796212","https://openalex.org/W2788496822","https://openalex.org/W2795282075","https://openalex.org/W2798665661","https://openalex.org/W2885593519","https://openalex.org/W2900260828","https://openalex.org/W2923014074","https://openalex.org/W2932893307","https://openalex.org/W2943552823","https://openalex.org/W2946609015","https://openalex.org/W2947380870","https://openalex.org/W2951013084","https://openalex.org/W2951286828","https://openalex.org/W2962736243","https://openalex.org/W2962843521","https://openalex.org/W2963159690","https://openalex.org/W2963341956","https://openalex.org/W2963351448","https://openalex.org/W2963394326","https://openalex.org/W2963748441","https://openalex.org/W2963846996","https://openalex.org/W2963961878","https://openalex.org/W2963969878","https://openalex.org/W2964044490","https://openalex.org/W2965373594","https://openalex.org/W2969601108","https://openalex.org/W2970019270","https://openalex.org/W2970379526","https://openalex.org/W2970597249","https://openalex.org/W2979826702","https://openalex.org/W2984256198","https://openalex.org/W2995793065","https://openalex.org/W2996908057","https://openalex.org/W3034584102","https://openalex.org/W3034831508","https://openalex.org/W3035139434","https://openalex.org/W3035608860","https://openalex.org/W3095992020","https://openalex.org/W3100350094","https://openalex.org/W3100895823","https://openalex.org/W3102259594","https://openalex.org/W3103934428","https://openalex.org/W3105302490","https://openalex.org/W3110511794","https://openalex.org/W3129973872","https://openalex.org/W3152911627","https://openalex.org/W3155431862","https://openalex.org/W3170037207","https://openalex.org/W3172908301","https://openalex.org/W3184352888","https://openalex.org/W3214527415","https://openalex.org/W4206288797","https://openalex.org/W4287588535","https://openalex.org/W4287824654","https://openalex.org/W4288348623","https://openalex.org/W4289375491","https://openalex.org/W4297667101","https://openalex.org/W4298289240","https://openalex.org/W4300996741","https://openalex.org/W4318142185","https://openalex.org/W4385573497","https://openalex.org/W4386576760"],"related_works":["https://openalex.org/W2502115930","https://openalex.org/W4246396837","https://openalex.org/W2482350142","https://openalex.org/W3176240006","https://openalex.org/W3126451824","https://openalex.org/W2180954594","https://openalex.org/W1561927205","https://openalex.org/W3191453585","https://openalex.org/W4297672492","https://openalex.org/W4288019534"],"abstract_inverted_index":{"Natural":[0],"language":[1],"inference":[2],"(NLI)":[3],"models":[4,54],"are":[5],"susceptible":[6],"to":[7,29,31,48,94,119],"learning":[8],"shortcuts,":[9],"i.e.":[10],"decision":[11],"rules":[12],"that":[13,127,144],"spuriously":[14],"correlate":[15],"with":[16,125],"the":[17,50,67,86,96,105,109,117,129,133],"label.":[18],"As":[19],"a":[20,45,76,80],"result,":[21],"they":[22],"achieve":[23],"high":[24,112,160],"in-distribution":[25,161],"performance,":[26],"but":[27],"fail":[28],"generalize":[30],"out-of-distribution":[32,60,154],"samples":[33],"where":[34,108],"such":[35],"correlations":[36],"do":[37],"not":[38],"hold.":[39],"In":[40],"this":[41,72],"paper,":[42],"we":[43,74],"present":[44],"training":[46],"method":[47,146],"reduce":[49],"reliance":[51],"of":[52,66,104],"NLI":[53,87,141],"on":[55,121,139,153],"shortcuts":[56,68,130],"and":[57,89],"improve":[58],"their":[59],"performance":[61],"without":[62],"assuming":[63],"prior":[64],"knowledge":[65],"being":[69,83],"targeted.":[70],"To":[71],"end,":[73],"propose":[75],"minimax":[77],"objective":[78],"between":[79],"learner":[81,110,118],"model":[82,92],"trained":[84],"for":[85],"task,":[88],"an":[90],"auxiliary":[91],"aiming":[93],"maximize":[95],"learner's":[97],"loss":[98],"by":[99],"up-weighting":[100],"examples":[101,124],"from":[102,132],"regions":[103],"input":[106],"space":[107],"incurs":[111],"losses.":[113],"This":[114],"process":[115],"incentivizes":[116],"focus":[120],"under-represented":[122],"\u201chard\u201d":[123],"patterns":[126],"contradict":[128],"learned":[131],"prevailing":[134],"\u201ceasy\u201d":[135],"examples.":[136],"Experimental":[137],"results":[138],"three":[140],"datasets":[142],"demonstrate":[143],"our":[145],"consistently":[147],"outperforms":[148],"other":[149],"robustness":[150],"enhancing":[151],"techniques":[152],"adversarial":[155],"test":[156],"sets,":[157],"while":[158],"maintaining":[159],"accuracy.":[162]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1}],"updated_date":"2026-04-28T14:05:53.105641","created_date":"2025-10-10T00:00:00"}