{"id":"W4415230413","doi":"10.36227/techrxiv.176053480.01524488/v1","title":"Semantic Pruning of Requirement Specifications: An NLP Framework for Redundancy Detection","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Software Engineering Techniques and Practices","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Redundancy (engineering); Word2vec; Software; Pruning","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":[],"consensus_categories":[],"category_scores_codex":[0.000558738,0.0001627306,0.000216544,0.0002062009,0.00007784824,0.0001531842,0.0009902061,0.0002312838,0.000008462384],"category_scores_gemma":[0.0003920121,0.0001675278,0.00009612471,0.0002434118,0.0000151024,0.0003010806,0.0004886807,0.0003270523,9.601142e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006688308,"about_ca_system_score_gemma":0.0001036891,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000665758,"about_ca_topic_score_gemma":0.000007303362,"domain_scores_codex":[0.9987755,0.0000455846,0.0003748721,0.0004782995,0.0001738052,0.0001519568],"domain_scores_gemma":[0.9977439,0.0004538094,0.0002992057,0.001277651,0.0001866893,0.00003873606],"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":[0.00001628002,0.000164218,0.00007089297,0.001140108,0.00009719748,9.372209e-7,0.0008298801,0.004126527,0.0006542135,0.8007274,0.0003467685,0.1918256],"study_design_scores_gemma":[0.0001256749,0.0003124266,0.0006128128,0.001560586,0.00007796021,0.000004075983,0.00005159667,0.2671421,0.05606302,0.6562229,0.01723836,0.0005885234],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0003224509,0.0002544151,0.9958404,0.0004548817,0.001101356,0.0006596798,0.000005888857,0.0008755248,0.0004854065],"genre_scores_gemma":[0.1304316,0.0001409766,0.8688862,0.00003416865,0.00009833161,0.0002167294,0.000007410609,0.000009006264,0.0001755547],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2630155,"threshold_uncertainty_score":0.6831587,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09976732863886802,"score_gpt":0.354254591948756,"score_spread":0.254487263309888,"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."}}