{"id":"https://openalex.org/W4409480950","doi":"https://doi.org/10.14428/esann/2025.es2025-150","title":"Evaluating Concept Discovery Methods for Sensitive Attributes in Language Models","display_name":"Evaluating Concept Discovery Methods for Sensitive Attributes in Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4409480950","doi":"https://doi.org/10.14428/esann/2025.es2025-150"},"language":"en","primary_location":{"id":"doi:10.14428/esann/2025.es2025-150","doi":"https://doi.org/10.14428/esann/2025.es2025-150","is_oa":true,"landing_page_url":"https://doi.org/10.14428/esann/2025.es2025-150","pdf_url":"https://doi.org/10.14428/esann/2025.es2025-150","source":{"id":"https://openalex.org/S7407087818","display_name":"European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ESANN 2025 proceesdings","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://doi.org/10.14428/esann/2025.es2025-150","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101154631","display_name":"Sarah Schr\u00f6der","orcid":"https://orcid.org/0000-0002-7954-3133","observed_orcids":["https://orcid.org/0000-0002-7954-3133"]},"institutions":[{"id":"https://openalex.org/I20121455","display_name":"Bielefeld University","ror":"https://ror.org/02hpadn98","country_code":"DE","type":"education","lineage":["https://openalex.org/I20121455"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Sarah Schr\u00f6der","raw_affiliation_strings":["- Bielefeld University -AG Machine Learning Inspiration 1 , 33615 Bielefeld - Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"- Bielefeld University -AG Machine Learning Inspiration 1 , 33615 Bielefeld - Germany","institution_ids":["https://openalex.org/I20121455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020392905","display_name":"Alexander Schulz","orcid":"https://orcid.org/0000-0002-0739-612X","observed_orcids":["https://orcid.org/0000-0002-0739-612X"]},"institutions":[{"id":"https://openalex.org/I20121455","display_name":"Bielefeld University","ror":"https://ror.org/02hpadn98","country_code":"DE","type":"education","lineage":["https://openalex.org/I20121455"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Alexander Schulz","raw_affiliation_strings":["- Bielefeld University -AG Machine Learning Inspiration 1 , 33615 Bielefeld - Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"- Bielefeld University -AG Machine Learning Inspiration 1 , 33615 Bielefeld - Germany","institution_ids":["https://openalex.org/I20121455"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5091180862","display_name":"Barbara Hammer","orcid":"https://orcid.org/0000-0002-0935-5591","observed_orcids":["https://orcid.org/0000-0002-0935-5591"]},"institutions":[{"id":"https://openalex.org/I20121455","display_name":"Bielefeld University","ror":"https://ror.org/02hpadn98","country_code":"DE","type":"education","lineage":["https://openalex.org/I20121455"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Barbara Hammer","raw_affiliation_strings":["- Bielefeld University -AG Machine Learning Inspiration 1 , 33615 Bielefeld - Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"- Bielefeld University -AG Machine Learning Inspiration 1 , 33615 Bielefeld - Germany","institution_ids":["https://openalex.org/I20121455"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I20121455"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.05466965,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"669","last_page":"674"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.8564000129699707,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.8564000129699707,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.1185000017285347,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.007300000172108412,"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/concept-discovery","display_name":"concept discovery","score":0.7839999794960022},{"id":"https://openalex.org/keywords/sensitive-attributes","display_name":"sensitive attributes","score":0.9670000076293945},{"id":"https://openalex.org/keywords/concept-of-identity","display_name":"concept of identity","score":0.6840000152587891},{"id":"https://openalex.org/keywords/language-models","display_name":"language models","score":0.5329999923706055},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.8190000057220459},{"id":"https://openalex.org/keywords/fairness","display_name":"fairness","score":0.7070000171661377},{"id":"https://openalex.org/keywords/label-sparsity","display_name":"label sparsity","score":0.8489999771118164},{"id":"https://openalex.org/keywords/pre-trained-language-models","display_name":"pre-trained language models","score":0.7789999842643738}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7364228367805481},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.41269540786743164},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3609282374382019},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3369324803352356}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.14428/esann/2025.es2025-150","doi":"https://doi.org/10.14428/esann/2025.es2025-150","is_oa":true,"landing_page_url":"https://doi.org/10.14428/esann/2025.es2025-150","pdf_url":"https://doi.org/10.14428/esann/2025.es2025-150","source":{"id":"https://openalex.org/S7407087818","display_name":"European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ESANN 2025 proceesdings","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.14428/esann/2025.es2025-150","doi":"https://doi.org/10.14428/esann/2025.es2025-150","is_oa":true,"landing_page_url":"https://doi.org/10.14428/esann/2025.es2025-150","pdf_url":"https://doi.org/10.14428/esann/2025.es2025-150","source":{"id":"https://openalex.org/S7407087818","display_name":"European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ESANN 2025 proceesdings","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.7307999730110168}],"sustainable_development_goals_aurora":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality education","score":0.4300000071525574}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4409480950.pdf","grobid_xml":"https://content.openalex.org/works/W4409480950.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"This":[0],"paper":[1],"examines":[2],"how":[3,31],"to":[4,32,73],"improve":[5],"interpretability":[6],"of":[7,13,36,53,63,87,94],"language":[8,41,89],"models":[9,90],"in":[10,39,56],"the":[11,21,34,47,60,85,92],"context":[12],"fairness.While":[14],"traditional":[15],"concept":[16],"learning":[17],"focuses":[18],"on":[19,91],"identifying":[20],"most":[22],"important":[23],"concepts":[24],"for":[25,66],"a":[26],"task,":[27],"this":[28,67],"study":[29],"explores":[30],"locate":[33],"representation":[35,93],"sensitive":[37,54,95],"attributes":[38,55],"pretrained":[40],"models.We":[42],"address":[43],"challenges":[44],"such":[45,75],"as":[46],"potential":[48,71],"low":[49],"importance":[50],"and":[51,59,84],"sparsity":[52],"training":[57],"data,":[58],"limited":[61],"amount":[62],"labeled":[64],"data":[65],"purpose.Our":[68],"experiments":[69],"evaluate":[70],"methods":[72],"obtain":[74],"identity":[76],"concepts,":[77],"considering":[78],"factors":[79],"like":[80],"label":[81],"sparsity,":[82],"generalizability,":[83],"influence":[86],"different":[88],"attributes.":[96]},"counts_by_year":[],"updated_date":"2026-10-06T10:28:23.643775","created_date":"2025-10-10T00:00:00"}