{"id":"W4405601877","doi":"10.1109/scam63643.2024.00017","title":"Enhancing Identifier Naming Through Multi-Mask Fine-Tuning of Language Models of Code","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Identifier; Computer science; Code (set theory); Programming language; Language model; Natural language processing; Artificial intelligence","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":[0.004033016,0.001274134,0.0007077755,0.001226128,0.0005302817,0.001788903,0.002066144,0.001193533,0.002401234],"category_scores_gemma":[0.02412901,0.0005230134,0.001163119,0.000733025,0.0009388077,0.005168797,0.002637065,0.001819171,0.001694886],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001160056,"about_ca_system_score_gemma":0.001940287,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002402189,"about_ca_topic_score_gemma":0.003845624,"domain_scores_codex":[0.9962122,0.001504293,0.0003281303,0.00108292,0.0007088108,0.0001635239],"domain_scores_gemma":[0.989087,0.005399891,0.001159758,0.002750569,0.001394886,0.0002078884],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001030075,0.0005557927,0.02105269,0.001052015,0.0001965203,0.0008090099,0.003087774,0.105055,0.1318269,0.02266542,0.01928568,0.6933832],"study_design_scores_gemma":[0.00006344634,0.0002100328,0.002055464,0.00005568756,0.000101738,0.0005553752,0.0002936159,0.9029576,0.06422557,0.01243645,0.01695717,0.00008793108],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09903084,0.0004157964,0.8607302,0.0005347864,0.0002191993,0.0003059558,0.0006331005,0.03551711,0.002613003],"genre_scores_gemma":[0.4982616,0.0002081929,0.4915424,0.0005198723,0.00005517878,0.0004101518,0.00217364,0.003487132,0.003341818],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004033016,"threshold_uncertainty_score":0.02132887,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02517924734871181,"score_gpt":0.315602785983944,"score_spread":0.2904235386352322,"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."}}