{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001900198,0.0002367053,0.0005645974,0.0005186997,0.0001117858,0.0002281268,0.0007509827,0.0002443315,0.000001770053],"category_scores_gemma":[0.000200403,0.0001949455,0.0001569211,0.0002947076,0.00003090008,0.000717853,0.0002783011,0.0005552086,1.246562e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002411104,"about_ca_system_score_gemma":0.0003860085,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001848243,"about_ca_topic_score_gemma":0.00001178635,"domain_scores_codex":[0.997422,0.0002933106,0.000997946,0.0003175752,0.0007942286,0.0001749865],"domain_scores_gemma":[0.9971175,0.00008586456,0.001450609,0.0005001607,0.000783913,0.00006192292],"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":[0.00008355915,0.001369154,0.001330823,0.02848728,0.001352637,0.0001143609,0.2436336,0.0308254,0.01997189,0.007867992,0.0003682553,0.6645951],"study_design_scores_gemma":[0.0005474083,0.00002408949,0.00007116879,0.002702354,0.0001332242,0.00005335223,0.001326633,0.9930893,0.0003230948,0.001533956,0.000005819385,0.0001896308],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04072061,0.0936079,0.8643935,0.00003172353,0.0005077301,0.0005882916,0.00002972772,0.0001034647,0.00001706975],"genre_scores_gemma":[0.5147206,0.00004401987,0.4851108,0.000008101849,0.00006074017,0.00001058106,0.00001702194,0.000007761954,0.00002029884],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9622639,"threshold_uncertainty_score":0.7949645,"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."}}