{"id":"W6948899260","doi":"10.5281/zenodo.11234583","title":"RUEG Corpus","year":2024,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"German; Heritage language; Variation (astronomy); Relevance (law); Language contact; Reading (process); Rule-based machine translation; Subject (documents)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"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":[0.0004266934,0.0002385547,0.0001960708,0.0002555182,0.0006236457,0.000753133,0.001364728,0.0002533159,0.007160247],"category_scores_gemma":[0.0003982166,0.0002525746,0.0001449364,0.0003272345,0.000153017,0.000007123273,0.002508272,0.0004053625,0.03159515],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000633594,"about_ca_system_score_gemma":0.000006712337,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002379974,"about_ca_topic_score_gemma":0.000001790492,"domain_scores_codex":[0.9982537,0.0001972825,0.0002585944,0.0006916812,0.0002898797,0.0003088871],"domain_scores_gemma":[0.9982992,0.000004769347,0.0001115712,0.001089865,0.0003649782,0.0001295837],"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":[0.0000210599,0.00006152259,4.787657e-8,0.0001067933,0.000120157,0.00003379862,0.0000101618,0.000001412899,0.0144176,0.00001314102,0.9761706,0.009043719],"study_design_scores_gemma":[0.0001074297,0.0002238094,0.000001957839,0.00003602136,0.00008828939,0.0001201627,0.00001503107,0.000007739168,0.005228176,0.00003691351,0.9938706,0.0002638523],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0001100068,0.0008573507,0.0003721521,0.0001444306,0.00009408002,0.0003660375,0.9736043,0.0004418659,0.02400978],"genre_scores_gemma":[0.000405086,0.001524407,0.00007253121,0.0002051486,0.0004314257,6.598687e-8,0.9939231,0.001281413,0.002156766],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.0244349,"threshold_uncertainty_score":0.9999927,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01516536557530488,"score_gpt":0.260243302567665,"score_spread":0.2450779369923602,"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."}}