{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005131579,0.002333336,0.001205447,0.0008680294,0.0007443764,0.001459138,0.003032609,0.002052417,0.0017327],"category_scores_gemma":[0.02558273,0.001160707,0.00116854,0.0007722221,0.002611266,0.004513196,0.002719926,0.006475526,0.0009123954],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00155053,"about_ca_system_score_gemma":0.002220168,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01233131,"about_ca_topic_score_gemma":0.01524455,"domain_scores_codex":[0.9983553,0.0004880661,0.0001073202,0.0005805776,0.0003108967,0.0001577487],"domain_scores_gemma":[0.984574,0.0114127,0.0006814866,0.001933173,0.001064785,0.0003338824],"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":[0.0002929523,0.000205271,0.004433915,0.0001798569,0.0001791297,0.0002723257,0.0003758568,0.7459595,0.007088678,0.01303722,0.002878514,0.2250968],"study_design_scores_gemma":[0.00001049771,0.00006371352,0.0003025131,0.00002294908,0.00001493852,0.00004371388,0.00001572653,0.9913459,0.002039175,0.005741098,0.0003862619,0.00001348249],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06139035,0.00148544,0.9308988,0.001437133,0.0001222971,0.00009754887,0.0001189702,0.002260334,0.002189141],"genre_scores_gemma":[0.8227046,0.001581395,0.1675836,0.001379155,0.0002022385,0.0002900994,0.0005694717,0.0005076122,0.005181841],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01233131,"threshold_uncertainty_score":0.02713865,"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."}}