{"id":"W4368755400","doi":"10.48550/arxiv.2305.02156","title":"Zero-Shot Listwise Document Reranking with a Large Language Model","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Pointwise; Ranking (information retrieval); Relevance (law); Information retrieval; Encoder; Pointwise mutual information; Task (project management); Natural language processing; Zero (linguistics); Language model; Identifier; Artificial intelligence; Mathematics; Mutual information; Linguistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.002118458,0.001355135,0.001589357,0.002167515,0.0006394098,0.001179021,0.001882601,0.0009930615,0.002664416],"category_scores_gemma":[0.005620625,0.0004308303,0.0007481659,0.001760627,0.0007004103,0.002671548,0.001087674,0.001441765,0.003282069],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006474553,"about_ca_system_score_gemma":0.001625554,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003594422,"about_ca_topic_score_gemma":0.01232905,"domain_scores_codex":[0.9981995,0.0006587368,0.0001123702,0.0004457587,0.0004611367,0.0001225595],"domain_scores_gemma":[0.996558,0.00134061,0.0002939478,0.0007815224,0.0009064482,0.0001195192],"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":[0.0003556906,0.0003915048,0.001445672,0.0003126936,0.0001500744,0.000156525,0.0001694262,0.04495361,0.03059524,0.006875541,0.01749253,0.8971015],"study_design_scores_gemma":[0.0001224485,0.0004643843,0.0007978909,0.00002487475,0.00007400239,0.0003204041,0.00008355299,0.9415787,0.03530902,0.01365307,0.007490567,0.00008104678],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02301589,0.000830726,0.966564,0.0001808147,0.0001354709,0.0001520056,0.0003169032,0.007481367,0.001322952],"genre_scores_gemma":[0.2517335,0.0004157924,0.7315146,0.0003447901,0.0002770378,0.0003158994,0.002818667,0.0006169323,0.01196274],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003594422,"threshold_uncertainty_score":0.01120359,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09702878345608794,"score_gpt":0.2197408233135964,"score_spread":0.1227120398575085,"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."}}