{"id":"W4416368800","doi":"10.48550/arxiv.2510.02534","title":"ZeroFalse: Improving Precision in Static Analysis with LLMs","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Static analysis; False positive paradox; Benchmark (surveying); Reliability (semiconductor); Precision and recall; Static program analysis; Software; Java","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01963303,0.004267813,0.002126629,0.008476991,0.001226458,0.006169197,0.005328496,0.002745902,0.003560972],"category_scores_gemma":[0.09834177,0.001839031,0.002909835,0.003733767,0.002269339,0.01179289,0.006959151,0.003669949,0.004154709],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002239176,"about_ca_system_score_gemma":0.006000287,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0096849,"about_ca_topic_score_gemma":0.01718901,"domain_scores_codex":[0.9754009,0.009066128,0.001986169,0.005185272,0.007438681,0.0009228755],"domain_scores_gemma":[0.9208629,0.04826175,0.004063853,0.01929611,0.006831915,0.0006835405],"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":[0.001521066,0.0006464875,0.04662479,0.002215484,0.0007684814,0.00051715,0.002020617,0.05840017,0.0304781,0.02099722,0.07420255,0.7616079],"study_design_scores_gemma":[0.0002603506,0.0004774698,0.005157977,0.000513325,0.0004136602,0.0007532289,0.0004889533,0.8277104,0.05576739,0.05998616,0.04818255,0.0002884951],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05452817,0.003431528,0.6480632,0.001777202,0.0003324427,0.0002604792,0.002946674,0.2838257,0.004834611],"genre_scores_gemma":[0.3609785,0.0008178641,0.6121281,0.001234514,0.0002079966,0.0002477739,0.008159427,0.012693,0.003532709],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01963303,"threshold_uncertainty_score":0.1038307,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03196078289550708,"score_gpt":0.2928788205595702,"score_spread":0.2609180376640631,"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."}}