{"id":"W6891699748","doi":"10.48448/xx6e-jf10","title":"AI Agents Learn to Trust","year":2023,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; McGill University","funders":"","keywords":"Task (project management); Reinforcement learning; Cornerstone; Work (physics); Social learning; Intelligent agent; Human intelligence","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","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001321094,0.0004529011,0.0004138708,0.002587031,0.0002694482,0.0002980647,0.002068127,0.0002343474,0.005414057],"category_scores_gemma":[0.0007143778,0.000424756,0.00007941924,0.005037984,0.001049488,0.0002225435,0.0006861833,0.0004530456,0.2130071],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000625429,"about_ca_system_score_gemma":0.0008959825,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00117617,"about_ca_topic_score_gemma":0.005037314,"domain_scores_codex":[0.9951347,0.00004781863,0.000331144,0.001321493,0.001904472,0.001260336],"domain_scores_gemma":[0.9977871,0.00003999335,0.0002092997,0.001174697,0.0001535959,0.000635361],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000003970138,0.00005130504,0.0002364214,0.00001250767,0.0000163625,0.00003239745,0.00009570898,0.0001479824,0.0005262442,0.0009821425,0.9902136,0.00768141],"study_design_scores_gemma":[0.000235835,0.00007558276,0.0004322295,0.0001795127,0.00002308029,0.000004815874,0.00009703134,0.002857164,0.00009747111,0.0002160679,0.9952237,0.000557481],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.0002145836,0.00007441834,0.0006967665,0.001812046,0.003259532,0.001120882,0.0004794363,0.005588767,0.9867536],"genre_scores_gemma":[0.002582977,0.00001577248,0.002774261,0.001986908,0.0007384591,0.00003505784,0.00003362632,0.002977422,0.9888555],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.2075931,"threshold_uncertainty_score":0.9998204,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04543487919129324,"score_gpt":0.3538621786037605,"score_spread":0.3084272994124672,"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."}}