{"id":"W4414113464","doi":"10.2139/ssrn.5472728","title":"From Law to Gherkin: A Human-Centred Quasi-Experiment on the Quality of LLM-Generated Behavioural Specifications from Food-Safety Regulations","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Food Supply Chain Traceability","field":"Agricultural and Biological Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa","funders":"","keywords":"CLARITY; Quality (philosophy); Agile software development; Relevance (law); Completeness (order theory); Software; Domain (mathematical analysis); Software requirements specification","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":"codex-gemma-dda1882f352a","candidate_categories":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001402628,0.0003962178,0.0005705225,0.00003087653,0.0007521138,0.0001570736,0.001297934,0.000324109,0.0006332902],"category_scores_gemma":[0.0001186076,0.0001703723,0.0004658377,0.0003412171,0.0001535964,0.00005690995,0.0003510154,0.002379918,0.00001303108],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00139748,"about_ca_system_score_gemma":0.0004817633,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.02222246,"about_ca_topic_score_gemma":0.07736914,"domain_scores_codex":[0.9952517,0.001038266,0.001134248,0.0007299189,0.0006859869,0.001159927],"domain_scores_gemma":[0.9977624,0.0006851116,0.0006062136,0.0004962229,0.0003015679,0.000148518],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001081766,0.006777877,0.009153033,0.00001653562,0.002003914,0.000001650155,0.004415402,0.001206168,0.2778599,0.683751,0.001400934,0.01233188],"study_design_scores_gemma":[0.001154861,0.003109513,0.3030989,0.0003452843,0.0003164292,0.000003489537,0.01762906,0.0001030888,0.02981405,0.6384367,0.004584575,0.001404052],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9777791,0.001208591,0.0000997803,0.01665319,0.000374782,0.0008742071,0.002528488,0.00005846818,0.0004234276],"genre_scores_gemma":[0.9973626,0.0001213682,0.0001154845,0.0001898282,0.0006048356,0.00006916541,0.001100477,0.000003639786,0.0004325693],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2939459,"threshold_uncertainty_score":0.9999216,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1130159363532756,"score_gpt":0.3045517953429158,"score_spread":0.1915358589896402,"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."}}