{"id":"W6947849807","doi":"10.48448/zkhk-xm79","title":"EvoGrad: An Online Platform for an Evolving Winograd Schema Challenge using Adversarial Human Perturbations","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Evolution and Paleontology Studies","field":"Earth and Planetary Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Brock University","funders":"","keywords":"Adversarial system; Schema (genetic algorithms); Task (project management); Metric (unit); Stability (learning theory); Measure (data warehouse)","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","sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0006790804,0.0003611499,0.0004017812,0.0007970942,0.002067434,0.0001334757,0.000924537,0.0002159944,0.007016832],"category_scores_gemma":[0.0001344969,0.0003304578,0.00009901716,0.0006307225,0.001112284,0.0007759934,0.00006564297,0.0003754731,0.00001614667],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005491211,"about_ca_system_score_gemma":0.0004222897,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004801734,"about_ca_topic_score_gemma":0.3452968,"domain_scores_codex":[0.9973404,0.00006810303,0.0003620181,0.0008752896,0.0006528005,0.0007013977],"domain_scores_gemma":[0.9986377,0.0001077236,0.0003007559,0.0005174995,0.0001428732,0.0002934074],"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":[0.0003658103,0.004115124,0.5780861,0.000685055,0.00113764,0.00009073488,0.01482605,0.01168252,0.0007281736,0.02575046,0.1536045,0.2089278],"study_design_scores_gemma":[0.002096164,0.00280688,0.04950687,0.0001268561,0.0002845116,0.00005556186,0.01097603,0.5626799,0.00000240992,0.002733682,0.3668543,0.001876838],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.07424125,0.01793086,0.005024468,0.001632401,0.01101609,0.005116833,0.008044616,0.002332879,0.8746606],"genre_scores_gemma":[0.7980481,0.0005358231,0.1202286,0.001336105,0.00715996,0.00004219798,0.01477537,0.0002962442,0.05757759],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.817083,"threshold_uncertainty_score":0.9999148,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1498071195975867,"score_gpt":0.3406532131337748,"score_spread":0.1908460935361881,"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."}}