{"id":"W4416536103","doi":"10.1038/s41597-025-05513-8","title":"Twenty years (2000–2020) of butterfly monitoring data across the contiguous United States","year":2025,"lang":"en","type":"article","venue":"Scientific Data","topic":"Lepidoptera: Biology and Taxonomy","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cascades (Canada)","funders":"U.S. Geological Survey; U.S. Fish and Wildlife Service; Israel Institute for Biological Research; U.S. Department of the Interior; National Science Foundation","keywords":"Butterfly; Taxonomy (biology); Biodiversity; Population; Taxonomic rank; Occupancy","routes":{"ca_aff":true,"ca_fund":false,"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":[0.001333598,0.0001044746,0.000117682,0.0000368627,0.000234386,0.0001174913,0.002773899,0.00008662325,0.00001263205],"category_scores_gemma":[0.0002792471,0.00007900928,0.00002477121,0.0003203041,0.0005677465,0.0000177084,0.002986845,0.0001259123,0.00001735913],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000006918018,"about_ca_system_score_gemma":0.0001375667,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001568379,"about_ca_topic_score_gemma":0.00007558229,"domain_scores_codex":[0.99861,0.0001054101,0.0002233887,0.0006933691,0.00009962537,0.0002681579],"domain_scores_gemma":[0.9960783,0.00004705752,0.00009396747,0.003659359,0.00008542751,0.00003589303],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003229952,0.0003941659,0.1147578,0.0001392427,0.0009176727,0.00002685831,0.0009317418,0.0001328272,0.04798467,0.00009637053,0.5093847,0.324911],"study_design_scores_gemma":[0.000266594,0.00002052131,0.00292219,0.0000268927,0.00002721428,0.000002576529,0.0004904215,0.0002593736,0.01371293,0.00003403078,0.9821371,0.0001001018],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.974581,0.001813094,0.001417061,0.0008006685,0.003491086,0.0002857709,0.01649561,0.00001915826,0.001096545],"genre_scores_gemma":[0.970234,0.0001222537,0.0003337331,0.0001433312,0.0001215917,0.000004746698,0.02722973,0.000005431977,0.001805161],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4727525,"threshold_uncertainty_score":0.5154641,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04054657016238197,"score_gpt":0.3212964209576983,"score_spread":0.2807498507953163,"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."}}