{"id":"W3083137311","doi":"10.1145/3409256.3409816","title":"Offline Evaluation without Gain","year":2020,"lang":"en","type":"article","venue":"","topic":"Information Retrieval and Search Behavior","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Compatibility (geochemistry); Computer science; Learning to rank; Ranking (information retrieval); Ideal solution; Information retrieval; Data mining; Artificial intelligence; Machine learning; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003266009,0.00004032175,0.00004715898,0.00002304335,0.00004120461,0.00007829216,0.0002677735,0.00001694015,0.0003113801],"category_scores_gemma":[0.0000753796,0.00003063097,0.0000208174,0.0002418855,0.000008283579,0.0004754022,0.00007672698,0.00005083137,0.0006123845],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001349025,"about_ca_system_score_gemma":0.00006686142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003949877,"about_ca_topic_score_gemma":5.959247e-7,"domain_scores_codex":[0.9992392,0.00003167866,0.0001155572,0.00009569819,0.0004229073,0.00009493943],"domain_scores_gemma":[0.9996018,0.00001064481,0.00002351602,0.0001211418,0.0001539308,0.0000889404],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002908444,0.00007540504,0.003734127,0.00002735143,0.00001353176,0.000005434842,0.0062377,0.00106464,0.006038654,0.2040474,0.02484385,0.7538829],"study_design_scores_gemma":[0.0002656437,0.00005772187,0.001479151,0.000001125237,0.000002060116,0.000001184901,0.00002280547,0.9858579,0.004736948,0.0002348558,0.007280794,0.00005977943],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01485359,0.0000077528,0.9430674,0.009622958,0.00007202164,0.0002058563,6.322676e-7,0.0001783347,0.03199148],"genre_scores_gemma":[0.9715185,0.000001272095,0.02352645,0.004312595,0.00004522534,0.000007553893,0.000005008211,0.000001805566,0.0005815572],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9847933,"threshold_uncertainty_score":0.7871167,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08723976651019504,"score_gpt":0.3242659391810598,"score_spread":0.2370261726708648,"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."}}