{"id":"W2952349219","doi":"10.18653/v1/p19-1166","title":"Understanding Undesirable Word Embedding Associations","year":2019,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":121,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Word (group theory); Debiasing; Association (psychology); Subspace topology; Computer science; Embedding; Projection (relational algebra); Natural language processing; Product (mathematics); Artificial intelligence; Association test; Word Association; Word embedding; Matrix decomposition; Speech recognition; Algorithm; Mathematics; Psychology; Social psychology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007496997,0.0007237821,0.0007916691,0.001651084,0.0009876014,0.002501348,0.000769073,0.001218098,0.0020761],"category_scores_gemma":[0.06115572,0.0005182474,0.0004475634,0.001637014,0.001822923,0.006437663,0.003746435,0.002058808,0.0007479257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005294199,"about_ca_system_score_gemma":0.0008697316,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001011909,"about_ca_topic_score_gemma":0.001594108,"domain_scores_codex":[0.9922993,0.003561903,0.0006462031,0.001663546,0.001465663,0.0003634956],"domain_scores_gemma":[0.9631541,0.02449203,0.003705146,0.004224978,0.003962374,0.000461401],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008557453,0.0002720597,0.1490989,0.0007598457,0.0003772876,0.000892394,0.007694384,0.04564612,0.02962867,0.2909563,0.01306869,0.4607496],"study_design_scores_gemma":[0.0000475443,0.0002382247,0.02517537,0.0001567881,0.0001219843,0.001423418,0.002453208,0.4637179,0.02001886,0.4683765,0.01817102,0.00009919974],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3426515,0.0008186165,0.6495171,0.00111165,0.0002086052,0.00006656774,0.0006687383,0.0006881474,0.004269043],"genre_scores_gemma":[0.9360655,0.0003229344,0.06086396,0.0002784043,0.0001337974,0.00009861207,0.0008437251,0.0002061911,0.001186959],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007496997,"threshold_uncertainty_score":0.03964841,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1460175943364272,"score_gpt":0.2942268895039399,"score_spread":0.1482092951675127,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}