{"id":"W4382239954","doi":"10.1609/aaai.v37i5.25757","title":"Better Peer Grading through Bayesian Inference","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Simons Institute for the Theory of Computing, University of California Berkeley; Alberta Machine Intelligence Institute; Compute Canada; Defense Advanced Research Projects Agency; University of British Columbia; Canadian Institute for Advanced Research","keywords":"Grading (engineering); Computer science; Inference; Interpretability; Rubric; Robustness (evolution); Machine learning; Bayesian probability; Probabilistic logic; Bayesian inference; Artificial intelligence; Mathematics; Mathematics education","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007906224,0.0002802676,0.0002923706,0.0001848073,0.0003288331,0.0004425043,0.002717026,0.0001049929,0.00008966231],"category_scores_gemma":[0.0008988078,0.000207372,0.0001561433,0.001701447,0.0002325258,0.0006267497,0.0006799343,0.000558158,0.0005526094],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003185914,"about_ca_system_score_gemma":0.00007701293,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006474485,"about_ca_topic_score_gemma":0.000003758551,"domain_scores_codex":[0.9973707,0.00002975674,0.0005246631,0.000633222,0.0008934595,0.0005482622],"domain_scores_gemma":[0.9983496,0.0001964936,0.0003220802,0.0004232827,0.0006119696,0.00009658508],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00001016211,0.00005585605,0.001131606,0.00003520657,0.00001521041,0.000001915031,0.002891711,0.0001588155,0.007070849,0.90272,0.001617257,0.08429137],"study_design_scores_gemma":[0.00003412014,0.0001627204,0.000657768,0.0002488861,0.00001049027,0.000006148432,0.0004493941,0.2988701,0.1499529,0.5475928,0.001623855,0.0003908471],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2635052,0.00003296515,0.5292467,0.1093719,0.004664289,0.001147168,0.00001732455,0.001982842,0.0900316],"genre_scores_gemma":[0.9897113,0.00002534972,0.007672605,0.0005145417,0.0001545846,0.00002052012,0.000001111004,0.00001849569,0.001881437],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7262062,"threshold_uncertainty_score":0.8456383,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07531137878603723,"score_gpt":0.3286441539662129,"score_spread":0.2533327751801757,"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."}}