{"id":"W6929258176","doi":"10.48448/zxex-ff58","title":"Carpe diem: On the Evaluation of World Knowledge in Lifelong Language Models","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Pipeline (software); Language model; Lifelong learning; Code (set theory); Benchmark (surveying); Work (physics); Dynamics (music); On Language","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":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.007676539,0.0003621334,0.0003798919,0.002991567,0.00006821194,0.0001088189,0.001221319,0.0001370032,0.001060471],"category_scores_gemma":[0.0005361687,0.000244677,0.00007895266,0.004681659,0.001085025,0.0001656987,0.0002992094,0.0005561832,0.004555237],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008006113,"about_ca_system_score_gemma":0.001462567,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001019766,"about_ca_topic_score_gemma":0.03542662,"domain_scores_codex":[0.9956451,0.0003207233,0.0004433782,0.000843381,0.002246701,0.0005007267],"domain_scores_gemma":[0.9980507,0.0002115496,0.0002906393,0.001066225,0.0002880096,0.00009282765],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003774817,0.0008120745,0.00006809811,0.0003637931,0.0001258075,0.00002665521,0.01266307,0.01714201,0.005020902,0.1256648,0.7954332,0.04264179],"study_design_scores_gemma":[0.0006937397,0.00008824687,0.00009661391,0.002947889,0.0002419169,0.000003242533,0.002098001,0.9539624,0.001115261,0.0141134,0.0239225,0.0007168293],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.003802848,0.005479951,0.00005106465,0.0002676285,0.0007680842,0.00113282,0.0001569838,0.0002253754,0.9881153],"genre_scores_gemma":[0.7288783,0.00001976788,0.0003631791,0.00008792238,0.0003996833,0.00009271098,0.00003377305,0.0006115078,0.2695132],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.9368203,"threshold_uncertainty_score":0.9998527,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1211401595697338,"score_gpt":0.3990946823107435,"score_spread":0.2779545227410096,"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."}}