{"id":"W4385572901","doi":"10.18653/v1/2022.emnlp-main.418","title":"TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Models","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":42,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea; Korea Advanced Institute of Science and Technology","keywords":"Perplexity; Computer science; Benchmark (surveying); Snapshot (computer storage); Machine learning; Lifelong learning; Artificial intelligence; Training set; Adaptability; Language model; Training (meteorology); Database","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009088253,0.004212366,0.001321082,0.003511036,0.001510059,0.002989924,0.00583454,0.003805049,0.004536683],"category_scores_gemma":[0.03792524,0.001052117,0.001599417,0.003398789,0.001319322,0.006921845,0.003051097,0.003699209,0.004167061],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002186814,"about_ca_system_score_gemma":0.002508282,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01958181,"about_ca_topic_score_gemma":0.03042904,"domain_scores_codex":[0.9930438,0.002878839,0.0008725356,0.001646454,0.001146522,0.0004118833],"domain_scores_gemma":[0.9821396,0.01046965,0.0006859752,0.003279008,0.002649206,0.000776584],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00169181,0.002768892,0.02219404,0.004677764,0.001788759,0.001076554,0.001045383,0.2751092,0.01138658,0.006062764,0.241377,0.4308212],"study_design_scores_gemma":[0.0003931463,0.0009473484,0.007159214,0.0002436758,0.0002001577,0.0005058722,0.0007177939,0.9219673,0.01890386,0.01055102,0.03823786,0.000172807],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5017239,0.01509957,0.2378759,0.004799785,0.003974248,0.002040063,0.08963805,0.110774,0.03407449],"genre_scores_gemma":[0.4904571,0.002288163,0.2925434,0.001311996,0.0003886354,0.0022,0.1965023,0.004973821,0.009334569],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01958181,"threshold_uncertainty_score":0.04806381,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09207660360202875,"score_gpt":0.3187674719700207,"score_spread":0.226690868367992,"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."}}