{"id":"W4396773582","doi":"10.1145/3664606","title":"Unveiling Code Pre-Trained Models: Investigating Syntax and Semantics Capacities","year":2024,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Software Engineering Research","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"National Research Foundation Singapore","keywords":"Computer science; Syntax; Abstract syntax tree; Abstract syntax; Programming language; Semantics (computer science); Syntax error; Artificial intelligence; Natural language processing; Code (set theory); Source code","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001034518,0.0002585097,0.0002968393,0.0004458509,0.0001544051,0.0001782104,0.0003795068,0.0001752276,0.000004529952],"category_scores_gemma":[0.002536092,0.0002621603,0.00005995013,0.0004510288,0.0001289149,0.0004048634,0.00004736915,0.0006489488,0.00000331773],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005041941,"about_ca_system_score_gemma":0.00006862812,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000358894,"about_ca_topic_score_gemma":0.00000427817,"domain_scores_codex":[0.9984082,0.0001383885,0.0002524591,0.0005672803,0.0002030948,0.0004306012],"domain_scores_gemma":[0.990473,0.008758945,0.00001936,0.0005074521,0.00004844028,0.0001927644],"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":[0.00001427245,0.00003839294,0.0001917686,0.001853129,0.0003743777,0.0001093156,0.01495256,0.7761028,0.006219144,0.0213549,0.00007028299,0.178719],"study_design_scores_gemma":[0.00023599,0.0001645324,0.0005001232,0.0004366683,0.00004604606,0.0004329773,0.0001590352,0.983555,0.006365551,0.006844719,0.0007801511,0.0004792241],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09211382,0.001815891,0.9031916,0.0003758662,0.0006598504,0.0001245788,0.00001480036,0.001700869,0.000002683457],"genre_scores_gemma":[0.3270372,0.0002026858,0.6724877,0.00004381928,0.00004320567,0.00004156076,0.000001373128,0.0000360527,0.0001063211],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2349234,"threshold_uncertainty_score":0.9999831,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1201043141820144,"score_gpt":0.3292689790807466,"score_spread":0.2091646648987322,"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."}}