{"id":"W4393594744","doi":"10.5281/zenodo.5139093","title":"TLMD: Tigrinya Language Modeling Dataset","year":2021,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Computer science; Natural language processing; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008551523,0.002788052,0.001340029,0.003089278,0.001672668,0.001907956,0.003141864,0.002286107,0.02979463],"category_scores_gemma":[0.003571525,0.0005851369,0.001718227,0.003524423,0.0006096552,0.002270376,0.002090011,0.002737378,0.05594197],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002372034,"about_ca_system_score_gemma":0.002949552,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04826804,"about_ca_topic_score_gemma":0.07358068,"domain_scores_codex":[0.99876,0.0002849968,0.0001288145,0.0003792957,0.0002688824,0.000178092],"domain_scores_gemma":[0.9988071,0.0003103955,0.00007643656,0.0003295231,0.0003348589,0.0001417264],"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":[0.0002682409,0.000224687,0.003514741,0.0007844582,0.00010416,0.0003184599,0.000153155,0.00222109,0.001388904,0.0009233472,0.9705673,0.01953134],"study_design_scores_gemma":[0.0005631245,0.0002280305,0.02268778,0.0004679767,0.0001657212,0.0008992777,0.001063076,0.02424265,0.004520606,0.003584182,0.9413468,0.0002308384],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.007673873,0.0004941049,0.001384992,0.0003607713,0.0001914685,0.0001389314,0.9793572,0.006151274,0.004247349],"genre_scores_gemma":[0.004874124,0.00007045476,0.001659477,0.00008582979,0.0000169834,0.0001589898,0.9916727,0.0001689088,0.001292415],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.04826804,"threshold_uncertainty_score":0.09967297,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03338556646129205,"score_gpt":0.2826125070333909,"score_spread":0.2492269405720988,"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."}}