{"id":"W7092497460","doi":"10.5281/zenodo.17390924","title":"War roudoù an dorneganed – Parallel Corpus (Breton Vannetais / French) – KorpusGW_FINAL_DATA v1.0","year":2025,"lang":"br","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"History of Computing Technologies","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Linguistic Association","funders":"","keywords":"Sentence; Section (typography); Metadata; Front (military); Corpus linguistics; Object (grammar)","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":[0.001301901,0.002673553,0.001366248,0.004007908,0.001959522,0.002156106,0.002428746,0.002813521,0.04332716],"category_scores_gemma":[0.004192854,0.0006691696,0.001166388,0.004311101,0.0007980202,0.001303233,0.002228998,0.001731959,0.0843186],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001635054,"about_ca_system_score_gemma":0.00250081,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.05845091,"about_ca_topic_score_gemma":0.09621741,"domain_scores_codex":[0.9985602,0.0003673923,0.0001229612,0.000439232,0.000300681,0.0002094652],"domain_scores_gemma":[0.9985459,0.0003319733,0.0000882231,0.0004468412,0.0004125167,0.0001746386],"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":[0.000170594,0.00007001476,0.0006354913,0.0006469366,0.00003236114,0.0001675262,0.0001710627,0.0002778301,0.0008630952,0.0008865588,0.9862204,0.009858169],"study_design_scores_gemma":[0.0002072127,0.00004209455,0.006352228,0.0002026361,0.00002351441,0.0002547017,0.000304402,0.0005055667,0.0009998811,0.0007096794,0.9903603,0.00003783035],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.003182544,0.0004789813,0.000491294,0.0002158718,0.0001657712,0.0001004618,0.9892045,0.001682393,0.004478343],"genre_scores_gemma":[0.001239649,0.00005434776,0.0007501636,0.00003760702,0.00001477474,0.0001362811,0.9959356,0.0001804232,0.001651157],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05845091,"threshold_uncertainty_score":0.1449438,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03619296995267359,"score_gpt":0.2515432038026172,"score_spread":0.2153502338499436,"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."}}