{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002183212,0.001470531,0.0006558885,0.0007813951,0.0003939005,0.001673307,0.001834915,0.001476672,0.001535535],"category_scores_gemma":[0.02110334,0.0006705432,0.001100366,0.0005623113,0.001281117,0.00574395,0.001964175,0.004182457,0.0006663609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001780075,"about_ca_system_score_gemma":0.001870855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01173908,"about_ca_topic_score_gemma":0.0118046,"domain_scores_codex":[0.9989538,0.0003099672,0.00005898156,0.0003637743,0.0001836192,0.0001297849],"domain_scores_gemma":[0.9887403,0.007333898,0.000491947,0.001743744,0.001314523,0.000375536],"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":[0.000807485,0.000490037,0.04222309,0.0004541068,0.000353724,0.0004448465,0.001400553,0.6726412,0.02609305,0.01130673,0.00717723,0.236608],"study_design_scores_gemma":[0.00001582917,0.00009972345,0.00185685,0.00002005331,0.00004178843,0.00003618636,0.00009231923,0.9869927,0.005748573,0.004454479,0.0006228741,0.00001863453],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8486233,0.000487641,0.141884,0.000913411,0.000109787,0.0001277704,0.0008036361,0.002950823,0.00409979],"genre_scores_gemma":[0.9653939,0.0001312156,0.0307589,0.0002742621,0.00001553243,0.0001255408,0.001701814,0.0002573213,0.0013415],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01173908,"threshold_uncertainty_score":0.02334148,"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."}}