{"id":"W4224940987","doi":"10.1145/3491102.3501999","title":"Human-AI Collaboration via Conditional Delegation: A Case Study of Content Moderation","year":2022,"lang":"en","type":"article","venue":"CHI Conference on Human Factors in Computing Systems","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":128,"is_retracted":false,"has_abstract":true,"ca_institutions":"Microsoft (Canada)","funders":"National Science Foundation","keywords":"Computer science; Delegation; Testbed; Scalability; Moderation; Benchmark (surveying); Data science; Knowledge management; Artificial intelligence; Machine learning; World Wide Web","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000852439,0.0002591631,0.0004175031,0.0004608227,0.001295709,0.0004173027,0.0006882922,0.00006254687,0.00003494234],"category_scores_gemma":[0.00003876963,0.0002804086,0.00005185379,0.0007148259,0.00006246426,0.000499041,0.0002761044,0.0003962344,0.000004531716],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004675535,"about_ca_system_score_gemma":0.0001697958,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003274152,"about_ca_topic_score_gemma":0.001167914,"domain_scores_codex":[0.9964155,0.0007806723,0.001057938,0.0006502286,0.0007877407,0.0003079218],"domain_scores_gemma":[0.9980369,0.0001541288,0.0005888063,0.0005869169,0.000556263,0.00007699247],"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":[0.00001277278,0.001544039,0.013937,0.00005615732,0.0000493072,0.0003966788,0.0451353,0.436721,0.005910361,0.4958194,0.0001398407,0.0002781024],"study_design_scores_gemma":[0.0007305442,0.001687052,0.002576479,0.00007493219,0.000009207359,0.0001391209,0.04583793,0.9433501,0.00185734,0.003306733,0.0000152919,0.0004152134],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8743539,0.00001141159,0.1234045,0.00005941291,0.0006657607,0.001095545,0.00001154586,0.0001146178,0.0002832522],"genre_scores_gemma":[0.9995582,1.188749e-7,0.00008345614,0.00005696958,0.00007197909,0.0001040241,0.00004628189,0.0000151617,0.00006387153],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5066292,"threshold_uncertainty_score":0.9999648,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1900154621451945,"score_gpt":0.369592466663248,"score_spread":0.1795770045180535,"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."}}