{"id":"W2963723885","doi":"","title":"Understanding the Origins of Bias in Word Embeddings","year":2019,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":58,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Word (group theory); Word embedding; Natural language processing; Embedding; Artificial intelligence; TRACE (psycholinguistics); Harm; Machine learning; Linguistics; Psychology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004814647,0.00009943401,0.0001167081,0.0002191396,0.00004086605,0.0001415154,0.001059983,0.00003967813,0.0001360728],"category_scores_gemma":[0.000202513,0.00006958447,0.00003815636,0.0002552242,0.00003100031,0.0002869657,0.0001980793,0.0005054864,0.00002636856],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002258279,"about_ca_system_score_gemma":0.00004925599,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001868276,"about_ca_topic_score_gemma":0.00003156546,"domain_scores_codex":[0.9990063,0.00008470783,0.0001920466,0.00022082,0.0003641515,0.0001319327],"domain_scores_gemma":[0.9993303,0.0002153007,0.0001709272,0.0001837385,0.00008125187,0.00001851802],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001243342,0.00001330868,0.01729636,0.00000612516,0.000007335012,0.000004496337,0.0006132139,0.0005096245,0.002001495,0.9747549,0.000008413047,0.004772237],"study_design_scores_gemma":[0.0004347606,0.0001299197,0.001468622,0.0004056588,0.000002261109,0.00001411582,0.0003304123,0.8239937,0.00343866,0.1682843,0.001266747,0.0002308921],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2328357,0.0002018027,0.7185431,0.009197863,0.0007855084,0.0003194141,0.000003651765,0.0004074719,0.03770554],"genre_scores_gemma":[0.9864337,0.00001325963,0.01292911,0.000174605,0.00001856909,0.000003634293,0.000003248478,0.000006297004,0.0004175359],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8234841,"threshold_uncertainty_score":0.2837572,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1021607067859051,"score_gpt":0.341356417713073,"score_spread":0.2391957109271678,"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."}}