{"id":"W4389163154","doi":"10.1145/3611643.3616244","title":"On the Usage of Continual Learning for Out-of-Distribution Generalization in Pre-trained Language Models of Code","year":2023,"lang":"en","type":"article","venue":"","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Fonds de recherche du Québec – Nature et technologies","keywords":"Computer science; Robustness (evolution); Encoder; Software; Artificial intelligence; Machine learning; Code (set theory); Language model; Source code; Programming language","routes":{"ca_aff":true,"ca_fund":true,"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":[0.000721943,0.0000535227,0.0001172205,0.00007426628,0.0000318298,0.00001285352,0.0002008252,0.00003215481,0.00001055564],"category_scores_gemma":[0.0002922236,0.00004085083,0.00004237508,0.0003555911,0.00002722942,0.000118282,0.00004129031,0.00005486235,0.000001585144],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000115945,"about_ca_system_score_gemma":0.00002485657,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003503294,"about_ca_topic_score_gemma":0.00003813395,"domain_scores_codex":[0.9992461,0.0001048693,0.0002510896,0.0001208984,0.0001687772,0.0001083113],"domain_scores_gemma":[0.9992724,0.0003385088,0.0001605454,0.0001343942,0.00007981927,0.0000143376],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003339002,0.00003984012,0.0002552929,0.0000345457,0.000009983612,5.636779e-7,0.01454996,0.442928,0.02584676,0.5103502,0.0004169795,0.005534524],"study_design_scores_gemma":[0.0003738637,0.00008449607,0.001995113,0.00002601815,0.000001935092,8.995985e-8,0.000597575,0.9849453,0.009674939,0.002115515,0.000140262,0.00004489627],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2601633,0.000005595132,0.7388175,0.0001494841,0.00005367646,0.0001584289,0.000007947003,0.0000406646,0.0006034423],"genre_scores_gemma":[0.9959328,0.000003561384,0.003009102,0.00003676508,0.000008392919,0.00001152781,0.00005028606,0.000004321316,0.0009432169],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7358083,"threshold_uncertainty_score":0.1665848,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04107671141226672,"score_gpt":0.300361861995498,"score_spread":0.2592851505832313,"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."}}