{"id":"W4407362568","doi":"10.1109/rsp64122.2024.10870993","title":"Advancing Formal Verification: Fine-Tuning LLMs for Translating Natural Language Requirements to CTL Specifications","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Mitacs","keywords":"Computer science; CTL*; Natural language; Programming language; Formal methods; Formal verification; Natural language processing; Chemistry","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":[0.005815406,0.001246948,0.0005736962,0.001649101,0.0006042804,0.002002308,0.00203152,0.00132122,0.003962184],"category_scores_gemma":[0.03639392,0.0005103428,0.001531511,0.0009149715,0.0009704826,0.003766602,0.003160595,0.001842837,0.002485283],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001511139,"about_ca_system_score_gemma":0.003562672,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006378981,"about_ca_topic_score_gemma":0.01284505,"domain_scores_codex":[0.9944136,0.002780986,0.0005261198,0.001019376,0.001054459,0.0002055971],"domain_scores_gemma":[0.9818512,0.01051623,0.0007849963,0.004853334,0.001704887,0.0002892952],"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":[0.0009976387,0.0008401212,0.01554317,0.002685646,0.0002976465,0.0007657788,0.002063662,0.1572294,0.06926198,0.02459454,0.04370942,0.6820111],"study_design_scores_gemma":[0.0001911907,0.00035404,0.001972446,0.0002199534,0.00009084491,0.000386329,0.0007051497,0.8861585,0.04580426,0.02738575,0.03665324,0.00007828976],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1291211,0.001098805,0.7781091,0.001495166,0.0002975992,0.0006429419,0.003648976,0.07815335,0.007432954],"genre_scores_gemma":[0.3579204,0.0003335296,0.6233119,0.0007548517,0.00003772894,0.0004714185,0.01119529,0.003646131,0.002328777],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006378981,"threshold_uncertainty_score":0.03075516,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03252903741174228,"score_gpt":0.3288965835567603,"score_spread":0.296367546145018,"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."}}