{"id":"W4408080426","doi":"10.1016/j.geoderma.2025.117223","title":"The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping","year":2025,"lang":"en","type":"article","venue":"Geoderma","topic":"Soil Geostatistics and Mapping","field":"Environmental Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ministry of Agriculture, Food and Rural Affairs; Dalhousie University","funders":"Social Sciences and Humanities Research Council of Canada; Agriculture and Agri-Food Canada; Natural Sciences and Engineering Research Council of Canada","keywords":"Leakage (economics); Environmental science; Soil science; Geology; Remote sensing","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005073337,0.0001063385,0.000129178,0.00004052169,0.0002817908,0.0001257971,0.0002348721,0.0000426742,0.000045813],"category_scores_gemma":[0.0002896419,0.00008554386,0.00002947687,0.0001706549,0.0001652809,0.0003492204,0.0005186115,0.00008403418,0.00003023244],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000542917,"about_ca_system_score_gemma":0.00003678233,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005798135,"about_ca_topic_score_gemma":0.0009991516,"domain_scores_codex":[0.9989059,0.00002745997,0.0003800651,0.0002973057,0.0001365823,0.0002526959],"domain_scores_gemma":[0.9989756,0.0003989221,0.0001175102,0.0004524322,0.0000216614,0.00003386538],"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":[0.0004162971,0.001073541,0.1618797,0.003173216,0.0003657455,0.01283761,0.01003922,0.1127457,0.01377935,0.009396074,0.06504446,0.6092491],"study_design_scores_gemma":[0.002224643,0.0000550266,0.0113091,0.0002625032,0.00004999691,0.00282342,0.001984879,0.9530727,0.001567028,0.02194604,0.004265844,0.0004388747],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9811903,0.00004172575,0.01509828,0.0005412091,0.0001424165,0.0006612544,0.0003710532,0.0000178137,0.00193593],"genre_scores_gemma":[0.9959808,0.000001973658,0.002689142,0.00007078731,0.00001342521,0.00005879644,0.0001446883,0.000008099709,0.001032266],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.840327,"threshold_uncertainty_score":0.3488377,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03913837256751391,"score_gpt":0.3139930845743008,"score_spread":0.2748547120067868,"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."}}