{"id":"W4391157645","doi":"10.48550/arxiv.2401.11314","title":"CodeAid: Evaluating a Classroom Deployment of an LLM-based Programming Assistant that Balances Student and Educator Needs","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Code (set theory); Software deployment; Transparency (behavior); Class (philosophy); Thematic analysis; Mathematics education; Software engineering; Multimedia; Programming language; Psychology; Qualitative research; Artificial intelligence; Sociology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006740528,0.0003136936,0.0004100266,0.0004166816,0.0001232472,0.0003197821,0.001081516,0.0001617551,0.00000402569],"category_scores_gemma":[0.00002689042,0.000317004,0.0001601387,0.000617376,0.0001159173,0.0001517781,0.001486572,0.0006390171,0.000005176388],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001771846,"about_ca_system_score_gemma":0.0006858429,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001295532,"about_ca_topic_score_gemma":0.00005522294,"domain_scores_codex":[0.9980083,0.0002204334,0.0002596414,0.0009094528,0.0002704556,0.0003317828],"domain_scores_gemma":[0.9984018,0.0001271669,0.0003401638,0.0007885455,0.000142594,0.0001997712],"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":[0.0001346156,0.003184908,0.3515529,0.002980147,0.001200742,0.0009002981,0.006669091,0.4451751,0.0003336811,0.1126743,0.0001156473,0.07507861],"study_design_scores_gemma":[0.0004492195,0.0004451903,0.003365319,0.0006198205,0.0002751817,0.000003607377,0.001719943,0.9854431,0.0001330783,0.006915832,0.0002022239,0.0004274863],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9614403,0.0003044369,0.03679011,0.0003934823,0.0004247387,0.0002709083,0.00001084071,0.0002497659,0.0001154003],"genre_scores_gemma":[0.9868479,0.00004217487,0.01263315,0.00004166127,0.00007187644,0.000002021139,0.00001173331,0.00002356866,0.0003259277],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5402681,"threshold_uncertainty_score":0.9999282,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0855630401474765,"score_gpt":0.2724213960874086,"score_spread":0.1868583559399321,"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."}}