{"id":"W4389473428","doi":"10.16995/dscn.9669","title":"Alignement sémantique et manque de données : l’apport des modèles de langue. Le cas du latin et du grec","year":2023,"lang":"fr","type":"article","venue":"Digital Studies / Le champ numérique","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Humanities; Philosophy","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001933756,0.0006301095,0.0006524032,0.0002329905,0.0005949688,0.0008918341,0.001039029,0.0002946923,0.000007588284],"category_scores_gemma":[0.0008961632,0.0006149266,0.000247574,0.0008745841,0.000608074,0.001827244,0.001897528,0.0005134069,0.00006106617],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004972942,"about_ca_system_score_gemma":0.0007008064,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004368492,"about_ca_topic_score_gemma":0.004898693,"domain_scores_codex":[0.9965188,0.0003622837,0.0006072035,0.0008836265,0.0004363332,0.001191782],"domain_scores_gemma":[0.9979217,0.0005165025,0.000291063,0.0005727325,0.0004557434,0.0002422415],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00005390057,0.0007581275,0.01643035,0.001746688,0.0006973504,0.004282366,0.7420769,0.0005532314,0.002146836,0.1354587,0.05736399,0.03843159],"study_design_scores_gemma":[0.001506961,0.0006825937,0.01001537,0.003450972,0.000005757175,0.001409386,0.2345264,0.01422883,0.04419882,0.6226105,0.06443246,0.002932008],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5544294,0.03495903,0.3101387,0.08274593,0.0007583944,0.0009587365,0.0002097681,0.00323848,0.01256155],"genre_scores_gemma":[0.9696411,0.00815314,0.01552438,0.001697834,0.0003427865,0.000256705,0.00007490418,0.00009844238,0.004210638],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5075505,"threshold_uncertainty_score":0.9996302,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05121940641463512,"score_gpt":0.3031767431746459,"score_spread":0.2519573367600108,"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."}}