{"id":"W7118051471","doi":"","title":"How Do Agentic AI Systems Address Performance Optimizations? A BERTopic-Based Analysis of Pull Requests","year":2025,"lang":"","type":"article","venue":"ArXiv.org","topic":"Software System Performance and Reliability","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Winnipeg; University of Windsor; University of Manitoba","funders":"","keywords":"Software; Focus (optics); Empirical research; Software system; Software development; Empirical evidence","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":[0.01101114,0.0005167444,0.0005347683,0.003526257,0.0009430827,0.003181382,0.00119018,0.0008270236,0.001029262],"category_scores_gemma":[0.07076483,0.0004335879,0.0004029688,0.001947874,0.0006723131,0.004432616,0.001391826,0.0008196087,0.0006565124],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001030905,"about_ca_system_score_gemma":0.001512184,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002078094,"about_ca_topic_score_gemma":0.002541859,"domain_scores_codex":[0.9932708,0.002784197,0.0005654999,0.0008759277,0.002093173,0.0004104076],"domain_scores_gemma":[0.9185011,0.0444795,0.01542492,0.007645861,0.01216213,0.001786451],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001320809,0.0007391133,0.4760718,0.001287833,0.0004469971,0.000661574,0.008868878,0.0543224,0.06166135,0.01487303,0.004651207,0.3750949],"study_design_scores_gemma":[0.00008798399,0.001504944,0.2727558,0.0001498367,0.0003981557,0.0009050142,0.006361267,0.6518263,0.03242311,0.01817077,0.01520871,0.0002081291],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9254993,0.0007224072,0.06640638,0.0005984925,0.00004026137,0.0002243007,0.0002072933,0.001394752,0.00490687],"genre_scores_gemma":[0.974434,0.0002196936,0.02383775,0.00007771295,0.00003631964,0.00009979529,0.0002250642,0.00011461,0.0009550303],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01101114,"threshold_uncertainty_score":0.05823314,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02569648182076357,"score_gpt":0.2723796907126856,"score_spread":0.246683208891922,"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."}}