{"id":"W7117477189","doi":"10.1145/3786771","title":"Assessing and Advancing Benchmarks for Evaluating Large Language Models in Software Engineering Tasks","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Software Engineering Research","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Popularity; Software quality assurance; Model-driven architecture; Software development; Coding (social sciences); Quality (philosophy); Software; Benchmark (surveying); Social software engineering","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":["metaresearch","metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.00251442,0.0003163923,0.0004593378,0.0008716071,0.0001719749,0.0001551698,0.0004688852,0.0002364165,0.000005419301],"category_scores_gemma":[0.01137996,0.0003553455,0.00008105001,0.0006415437,0.00002423685,0.0005450296,0.00008839968,0.0006496983,4.820315e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001344542,"about_ca_system_score_gemma":0.0000881683,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003445131,"about_ca_topic_score_gemma":0.00000923111,"domain_scores_codex":[0.9978436,0.0001446729,0.0003763087,0.0006984881,0.0001903807,0.0007465046],"domain_scores_gemma":[0.9854408,0.01367178,0.0000430449,0.0006220002,0.00008399889,0.0001384092],"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":[0.00002734721,0.00006594922,0.001322504,0.0007732847,0.0001058047,0.00002303362,0.001844179,0.7534376,0.002460918,0.002097492,0.00002076434,0.2378211],"study_design_scores_gemma":[0.002028344,0.0002143082,0.009188664,0.0007415403,0.00005829238,0.00006969034,0.0003258265,0.9809336,0.003700092,0.001377782,0.0006276167,0.0007342161],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08182426,0.00124761,0.9153393,0.0001358307,0.0005430578,0.0003406783,0.00001156707,0.0005555677,0.000002141599],"genre_scores_gemma":[0.2112775,0.00005483853,0.788303,0.00007044262,0.00002302298,0.0001955321,0.000006266637,0.00003379578,0.00003568877],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2370868,"threshold_uncertainty_score":0.9998899,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07511031672640531,"score_gpt":0.3920066529526886,"score_spread":0.3168963362262833,"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."}}