{"id":"W4293097939","doi":"10.1145/3478432.3499048","title":"Using Deep Learning to Localize Errors in Student Code Submissions","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 53rd ACM Technical Symposium on Computer Science Education V. 2","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Python (programming language); Computer science; Deep learning; Transfer of learning; Artificial intelligence; Coding (social sciences); Machine learning; Metric (unit); Natural language processing; Programming language; Statistics","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":["open_science"],"consensus_categories":[],"category_scores_codex":[0.001664832,0.0001819356,0.0002264294,0.0005704296,0.0008830035,0.0002771457,0.006236582,0.00004338823,0.000007605764],"category_scores_gemma":[0.0004327546,0.0001497878,0.00009662174,0.004352981,0.0001922397,0.0003953259,0.004765302,0.0007047976,0.000004945804],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005551162,"about_ca_system_score_gemma":0.0005987115,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003012788,"about_ca_topic_score_gemma":0.000002337835,"domain_scores_codex":[0.9970202,0.00004496333,0.0004573188,0.0007586277,0.001264093,0.0004547657],"domain_scores_gemma":[0.9985207,0.0001002338,0.0002986382,0.0005594942,0.0003053392,0.0002156034],"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":[0.00003617659,0.003788163,0.0903426,0.00007938087,0.0000168857,0.000002269648,0.009717381,0.5256284,0.1966354,0.1281487,0.002484953,0.0431198],"study_design_scores_gemma":[0.0004849904,0.001077993,0.0552158,0.0004111087,0.00002388863,0.0001021027,0.001457201,0.9122957,0.008367197,0.01031108,0.009407504,0.00084544],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9177741,0.00002157065,0.02310854,0.05586466,0.001588873,0.0005783289,0.000001031256,0.000260539,0.0008022917],"genre_scores_gemma":[0.9029379,0.00000273032,0.09584336,0.0009131257,0.00009570268,0.00003111039,2.988428e-7,0.00001148654,0.0001642706],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3866673,"threshold_uncertainty_score":0.9991401,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02324682460761803,"score_gpt":0.3317174811870447,"score_spread":0.3084706565794267,"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."}}