{"id":"W4401200995","doi":"10.48550/arxiv.2407.18940","title":"LitSearch: A Retrieval Benchmark for Scientific Literature Search","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Semantic Web and Ontologies","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute for Catastrophic Loss Reduction; National Science Foundation","keywords":"Benchmark (surveying); Information retrieval; Computer science; Scientific literature; Data science; Geography; Biology; Cartography; Paleontology","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01176248,0.00212703,0.001443969,0.02262137,0.001627465,0.003551015,0.00245005,0.00254108,0.007575789],"category_scores_gemma":[0.06257696,0.000507835,0.001416129,0.01530803,0.0009494147,0.005176493,0.003577145,0.001265831,0.008172857],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002117197,"about_ca_system_score_gemma":0.003849952,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006469675,"about_ca_topic_score_gemma":0.01215959,"domain_scores_codex":[0.989316,0.004297523,0.002207902,0.001177715,0.00264094,0.0003600175],"domain_scores_gemma":[0.9676444,0.01872533,0.002074718,0.004244331,0.006228523,0.001082683],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001473126,0.001041679,0.01102721,0.02351359,0.0008847167,0.0006345296,0.001657921,0.02306626,0.02247682,0.01168096,0.4606257,0.4419175],"study_design_scores_gemma":[0.001931888,0.003083284,0.03234949,0.002670037,0.0008823879,0.002875312,0.002627251,0.219609,0.05543976,0.03828269,0.6396486,0.0006002426],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.1847364,0.06186186,0.2079415,0.007372284,0.002245199,0.005924559,0.3451386,0.1306673,0.05411211],"genre_scores_gemma":[0.1677641,0.006531709,0.2956322,0.001374782,0.0005398119,0.003429776,0.5149957,0.00320142,0.006530607],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9882375,"threshold_uncertainty_score":0.06220669,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.136358188139202,"score_gpt":0.2335341740633809,"score_spread":0.09717598592417886,"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."}}