{"id":"W4412827352","doi":"10.1145/3744340","title":"AcTracer: Active Testing of Large Language Model via Multi-Stage Sampling","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Sampling (signal processing)","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":[0.0005439415,0.0001501999,0.0002635001,0.0002623359,0.00009780046,0.00001465663,0.0003666774,0.0001255353,0.000002755388],"category_scores_gemma":[0.001121659,0.0001581727,0.00005849676,0.0003060143,0.0000183366,0.0001434388,0.00003724211,0.0003249067,6.214561e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003194217,"about_ca_system_score_gemma":0.00004757191,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004494567,"about_ca_topic_score_gemma":0.000007977263,"domain_scores_codex":[0.9990005,0.00008548054,0.0002264315,0.0003460532,0.00007786077,0.0002636588],"domain_scores_gemma":[0.9972041,0.002102636,0.00005116201,0.0005341039,0.00005510049,0.00005295203],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001000735,0.00005432821,0.00006551527,0.0001082882,0.00005717835,0.000002790435,0.00171917,0.6972253,0.01898883,0.001423484,5.230385e-7,0.2803445],"study_design_scores_gemma":[0.0004015915,0.0000317312,0.0003487848,0.00006318833,0.00002333869,0.000008115766,0.0001062826,0.9715452,0.02686995,0.0003805174,0.00007592082,0.0001453897],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03291576,0.0001772235,0.9661418,0.00006419398,0.0002628689,0.00009893984,0.00001511562,0.0003157514,0.000008298252],"genre_scores_gemma":[0.2164716,0.00001157277,0.7832792,0.00006853999,0.000007960021,0.00001573471,7.632264e-7,0.000009642969,0.000134945],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2801992,"threshold_uncertainty_score":0.6450095,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1514059113826057,"score_gpt":0.3741478920576587,"score_spread":0.222741980675053,"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."}}