{"id":"W4324302602","doi":"10.48550/arxiv.2303.06233","title":"Model-Agnostic Syntactical Information for Pre-Trained Programming Language Models","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Software Engineering Research","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Automatic summarization; Abstract syntax tree; Source code; Language model; Security token; Transformer; Artificial intelligence; Scratch; Syntax; Programming language; Code (set theory); Natural language processing","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001246222,0.002522247,0.0008479636,0.001208049,0.0004290821,0.001154994,0.002782021,0.001334684,0.004695896],"category_scores_gemma":[0.006063925,0.0009336535,0.001886915,0.0009010736,0.0005533071,0.003384477,0.001454507,0.004601572,0.004009158],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00145101,"about_ca_system_score_gemma":0.001923378,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006782562,"about_ca_topic_score_gemma":0.01392083,"domain_scores_codex":[0.9989684,0.0002510735,0.00006478908,0.0004586108,0.000151117,0.000106102],"domain_scores_gemma":[0.9974366,0.001333221,0.0001253763,0.0004940866,0.0005289527,0.0000818387],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004274234,0.0003502117,0.004148512,0.0006058739,0.0002689468,0.0002444719,0.0002291973,0.3783984,0.02360365,0.003901216,0.02539617,0.5624259],"study_design_scores_gemma":[0.00002202371,0.00006300955,0.0003723262,0.00003059078,0.00004977615,0.00004016207,0.00002902069,0.9864125,0.007250076,0.003209192,0.002502603,0.00001873138],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08585504,0.001431551,0.8301272,0.0008987715,0.0003624698,0.0003269381,0.003093536,0.07168156,0.006222928],"genre_scores_gemma":[0.5409276,0.0007403505,0.4199628,0.001384525,0.000147483,0.001007705,0.02046619,0.004233917,0.0111294],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006782562,"threshold_uncertainty_score":0.01570934,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09852245320196315,"score_gpt":0.2433023073693213,"score_spread":0.1447798541673582,"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."}}