{"id":"W3040460157","doi":"10.48550/arxiv.2007.00700","title":"Response by the Montreal AI Ethics Institute to the Santa Clara Principles on Transparency and Accountability in Online Content Moderation","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Moderation; Transparency (behavior); Accountability; Content (measure theory); Political science; Psychology; Law; Social psychology; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000980903,0.0002346557,0.0002087597,0.00009367762,0.0003156583,0.0001926186,0.001064766,0.000268,0.000001838007],"category_scores_gemma":[0.0001990699,0.0001747375,0.0000775125,0.0004026695,0.0001151112,0.0002833267,0.0005360821,0.001467346,0.000009287771],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002085729,"about_ca_system_score_gemma":0.0002760458,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003192936,"about_ca_topic_score_gemma":0.01113752,"domain_scores_codex":[0.997867,0.0006499922,0.0002261752,0.0008755379,0.0001629357,0.0002183615],"domain_scores_gemma":[0.9985278,0.0002766915,0.0001062922,0.0008698387,0.0001074602,0.0001119198],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.005113766,0.0005619918,0.002958031,0.0001087616,0.000128569,0.0002258827,0.01862236,0.9150425,0.001468157,0.04630687,0.00096119,0.008501954],"study_design_scores_gemma":[0.001019302,0.0003985209,0.04563304,0.0001944999,0.0000518419,0.000007936588,0.0006597184,0.9354815,0.0008775924,0.009074785,0.006035392,0.0005658804],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7357798,0.00003983466,0.2479426,0.01523829,0.000234986,0.0005507598,0.00005444972,0.00007443411,0.00008493585],"genre_scores_gemma":[0.9971706,0.0002021742,0.0001897672,0.002224677,0.00003127184,0.000003926012,0.0000192117,0.000007883174,0.000150518],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2613908,"threshold_uncertainty_score":0.7125589,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2094321816893306,"score_gpt":0.2477945244998991,"score_spread":0.03836234281056847,"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."}}