{"id":"W4411868550","doi":"10.1016/j.jss.2026.112941","title":"Automatic Translation of Natural Language Requirements into Ctl Specifications Using Large Language Models: A Multi-Approach Evaluation ⋆","year":2025,"lang":"en","type":"preprint","venue":"Journal of Systems and Software","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"Mitacs","keywords":"Computer science; Translation (biology); Natural language processing; CTL*; Natural language; Programming language; Universal Networking Language; Artificial intelligence; Machine translation; Programming language specification; Software engineering; Comprehension approach","routes":{"ca_aff":true,"ca_fund":true,"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":[0.003319047,0.0008958926,0.0007872217,0.001072521,0.0006451731,0.002210196,0.001688801,0.001437922,0.008814245],"category_scores_gemma":[0.0134292,0.0005892846,0.001397859,0.001033029,0.0008344901,0.002709272,0.002145786,0.001427004,0.002361237],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001141331,"about_ca_system_score_gemma":0.00291553,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005027529,"about_ca_topic_score_gemma":0.005216999,"domain_scores_codex":[0.9953446,0.002190219,0.0002760511,0.000616469,0.001239604,0.0003330235],"domain_scores_gemma":[0.9844326,0.009291434,0.0006677785,0.002158806,0.003144302,0.0003052337],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003953678,0.002570635,0.007699775,0.00165873,0.0003524923,0.001306799,0.001840989,0.1825289,0.1005593,0.04138549,0.014595,0.6415483],"study_design_scores_gemma":[0.0003852809,0.000519068,0.00120441,0.00005041039,0.0001266694,0.0002812806,0.0006540961,0.9260546,0.05306206,0.01307761,0.004539385,0.00004505827],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2448055,0.0003600575,0.7317935,0.0006313364,0.000110602,0.0006904698,0.001056661,0.01142705,0.009124755],"genre_scores_gemma":[0.5281615,0.000166898,0.463558,0.000195403,0.00002508992,0.0002135963,0.002563565,0.001387728,0.003728193],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008814245,"threshold_uncertainty_score":0.0294866,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1028895751960369,"score_gpt":0.3616466353642591,"score_spread":0.2587570601682221,"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."}}