{"id":"W4408362526","doi":"10.1007/s11390-024-5036-x","title":"Caging AI","year":2025,"lang":"en","type":"article","venue":"Journal of Computer Science and Technology","topic":"Computability, Logic, AI Algorithms","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Artificial 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":[],"consensus_categories":[],"category_scores_codex":[0.001480735,0.0001242832,0.0002826534,0.001764323,0.0002484911,0.0003212693,0.002570456,0.00008213361,0.000001210398],"category_scores_gemma":[0.0001462288,0.0001005058,0.00004485478,0.003827731,0.0007749202,0.0009420995,0.001531189,0.0003828358,0.000002486741],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001003579,"about_ca_system_score_gemma":0.0006989912,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000286014,"about_ca_topic_score_gemma":0.000001210765,"domain_scores_codex":[0.9983972,0.00002872694,0.0004157485,0.0003654458,0.0004449079,0.0003479932],"domain_scores_gemma":[0.9980077,0.00009336011,0.0001919211,0.0005192509,0.001097422,0.00009030234],"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":[0.000001700991,0.00006106658,0.003348401,0.00000967753,0.00001287326,0.00005822038,0.0001179503,0.0000844749,0.001042269,0.1617335,0.0008230427,0.8327068],"study_design_scores_gemma":[0.0009605452,0.0007740995,0.01116623,0.0001447397,0.00001711306,0.001916251,0.00004406426,0.509442,0.01142586,0.442695,0.02106033,0.000353804],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05247243,0.0004369284,0.9175858,0.02762164,0.00144675,0.00006472954,9.302929e-8,0.000104146,0.000267413],"genre_scores_gemma":[0.7652669,0.00002266479,0.2333113,0.001304675,0.00007378831,9.778345e-7,1.326063e-8,0.000002134129,0.00001756733],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.832353,"threshold_uncertainty_score":0.477659,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00567773826158046,"score_gpt":0.255621987237542,"score_spread":0.2499442489759615,"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."}}