{"id":"W4402335544","doi":"10.1101/2024.09.06.24313186","title":"Loon Lens 1.0 Validation: Agentic AI for Title and Abstract Screening in Systematic Literature Reviews","year":2024,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Systematic review; Lens (geology); Through-the-lens metering; Psychology; Data science; Political science; Computer science; MEDLINE; Law; Physics; Optics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001897662,0.00007679777,0.0002526784,0.000110053,0.00005454853,0.0001783933,0.00008859616,0.00008790711,0.0001214783],"category_scores_gemma":[0.0003499358,0.00006204064,0.000110308,0.0001819105,0.00001975914,0.00003164867,0.00006324811,0.0001880998,0.00006519591],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001792466,"about_ca_system_score_gemma":0.00004104462,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002755257,"about_ca_topic_score_gemma":0.00006268789,"domain_scores_codex":[0.9991726,0.000186272,0.0002433337,0.0001843474,0.0001342973,0.00007913947],"domain_scores_gemma":[0.9995507,0.0001965367,0.0000879369,0.00009140577,0.0000488913,0.00002452296],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00002852834,0.0002430918,0.01637535,0.5132008,0.001069392,0.0001625837,0.07223674,0.001597566,0.0001603308,0.2025949,0.05062964,0.1417011],"study_design_scores_gemma":[0.0004066439,0.0000275049,0.01410826,0.1261674,0.001411246,0.000006665292,0.0007198405,0.01743001,0.00005597385,0.4167733,0.42156,0.001333195],"study_design_candidate":"systematic_review","study_design_consensus":null,"genre_codex":"review","genre_gemma":"empirical","genre_scores_codex":[0.1455189,0.5717602,0.07320799,0.01969224,0.009597339,0.01165428,0.000160062,0.0005083057,0.1679007],"genre_scores_gemma":[0.9318455,0.005358307,0.02348298,0.000388306,0.001770277,0.0003959179,0.0001660347,0.00004817353,0.03654454],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7863265,"threshold_uncertainty_score":0.2529944,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1130685049969929,"score_gpt":0.4176119119838094,"score_spread":0.3045434069868165,"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."}}