{"id":"W2997847174","doi":"10.1609/aaai.v34i05.6430","title":"TreeGen: A Tree-Based Transformer Architecture for Code Generation","year":2020,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Software Engineering Research","field":"Computer Science","cited_by":163,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"National Key Research and Development Program of China; National Natural Science Foundation of China","keywords":"Computer science; Python (programming language); Code generation; Parsing; Programming language; Parse tree; Redundant code; Encoder; Source code; Abstract syntax tree; Artificial intelligence; Theoretical computer science; Operating system","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":[],"consensus_categories":[],"category_scores_codex":[0.0003136228,0.0002061448,0.0002217778,0.0001027321,0.0001412226,0.0002013577,0.001751309,0.00008624027,0.00002315607],"category_scores_gemma":[0.001029513,0.0001568569,0.0001610693,0.0006628483,0.0001223808,0.0002031606,0.00007002766,0.0002742306,0.00002318742],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003553077,"about_ca_system_score_gemma":0.000153001,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000672412,"about_ca_topic_score_gemma":0.00001325209,"domain_scores_codex":[0.9982549,0.00001170123,0.0003773909,0.0005055133,0.0004986061,0.0003519043],"domain_scores_gemma":[0.9987969,0.0002420635,0.0001263306,0.0002180818,0.000474627,0.0001419697],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001988719,0.0001329079,0.0001826823,0.0001844709,0.00002775844,5.255069e-7,0.002801417,0.005029698,0.4683053,0.2980646,0.0008297429,0.2242419],"study_design_scores_gemma":[0.00004441503,0.000293637,0.00004677313,0.00004083313,0.00000587341,5.98926e-7,0.00003440676,0.3645267,0.6286016,0.005895407,0.0003755529,0.0001342847],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03614493,0.00001818581,0.939642,0.02260954,0.000193949,0.0007520635,0.00001613488,0.0001747667,0.0004484175],"genre_scores_gemma":[0.9795436,0.00000457793,0.01953337,0.0006030531,0.0001542487,0.0001006844,0.000001499303,0.00001826568,0.00004066596],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9433987,"threshold_uncertainty_score":0.6396439,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1211806594404359,"score_gpt":0.3057672945531414,"score_spread":0.1845866351127055,"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."}}