{"id":"W4313137049","doi":"10.1145/3524610.3527921","title":"An exploratory study on code attention in BERT","year":2022,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Automatic summarization; Transformer; Artificial intelligence; Programming language; Natural language processing; Natural language; Source code; Engineering","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":[0.001239585,0.0002536072,0.0002453678,0.001395526,0.00122279,0.002032706,0.0007572312,0.0009186968,0.009544089],"category_scores_gemma":[0.02486609,0.0002180129,0.0001660141,0.002364805,0.001334147,0.004180683,0.001404628,0.00115025,0.0009827965],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001879625,"about_ca_system_score_gemma":0.0008992985,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01819973,"about_ca_topic_score_gemma":0.01755883,"domain_scores_codex":[0.999056,0.0003033748,0.00002692767,0.0001805382,0.0002911976,0.0001418919],"domain_scores_gemma":[0.9891083,0.00786043,0.0008282546,0.0005400022,0.001140224,0.0005227728],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009790327,0.0009288813,0.3864418,0.0009445043,0.00009639982,0.00231341,0.117474,0.005664357,0.01084465,0.07331169,0.02403065,0.3769706],"study_design_scores_gemma":[0.00007256446,0.0006752252,0.586931,0.0005549417,0.0001190265,0.002629329,0.09228006,0.06545902,0.006902393,0.05627556,0.1879564,0.0001444313],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9440353,0.0009734184,0.005645834,0.001727525,0.00002869074,0.00006512665,0.0003002069,0.0001852916,0.04703855],"genre_scores_gemma":[0.9926865,0.0002171465,0.001292986,0.0003247352,0.00001770366,0.00002837933,0.0002931468,0.0000846148,0.00505481],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01819973,"threshold_uncertainty_score":0.03618759,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04139991490038435,"score_gpt":0.3098199987926032,"score_spread":0.2684200838922188,"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."}}