{"id":"W4327671617","doi":"10.48550/arxiv.2303.08518","title":"UPRISE: Universal Prompt Retrieval for Improving Zero-Shot Evaluation","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Task (project management); Computer science; Generalization; Set (abstract data type); Zero (linguistics); Shot (pellet); Labrador Retriever; Filling-in; Code (set theory); Computer security; Artificial intelligence; Medicine; Engineering; Programming language; Pathology; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009021903,0.003516371,0.002116669,0.001990672,0.0008540001,0.003070959,0.003569293,0.002927127,0.01384608],"category_scores_gemma":[0.03834898,0.0006851162,0.001110732,0.000930618,0.0009320856,0.006322544,0.005428988,0.003299338,0.008823624],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001296382,"about_ca_system_score_gemma":0.001704864,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003754098,"about_ca_topic_score_gemma":0.006853539,"domain_scores_codex":[0.993708,0.002879508,0.0004509786,0.001406259,0.001123256,0.0004319468],"domain_scores_gemma":[0.9904198,0.005580602,0.0002720621,0.00189817,0.001314419,0.0005149333],"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":[0.00295491,0.00068554,0.004134479,0.002162618,0.0005202582,0.0004198238,0.0009662327,0.03345601,0.04084599,0.006628782,0.1206525,0.7865729],"study_design_scores_gemma":[0.0007365589,0.001558045,0.00363955,0.0002677935,0.0002316777,0.0006361866,0.0005721376,0.8770415,0.05099869,0.03181279,0.03224443,0.0002606413],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06785131,0.006218413,0.7283207,0.0008276002,0.001474146,0.0009532609,0.005590357,0.1769656,0.01179864],"genre_scores_gemma":[0.5100046,0.0009647292,0.4410125,0.00167617,0.0005674648,0.00131274,0.02001389,0.01178566,0.01266222],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01384608,"threshold_uncertainty_score":0.04771298,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2033081645517611,"score_gpt":0.2405385146988309,"score_spread":0.03723035014706982,"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."}}