{"id":"W4408418184","doi":"10.5194/egusphere-2025-846-v1","title":"A novel hybrid fine-tuning method for supercharging deep learning model development for hydrological prediction","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Hydrological Forecasting Using AI","field":"Environmental Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Resources Canada","keywords":"Deep learning; Computer science; Artificial intelligence; Development (topology); Hybrid learning; Machine learning; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002149982,0.0005014013,0.0005854537,0.000114041,0.0006572605,0.00009363695,0.0005841409,0.0004682496,0.0002232086],"category_scores_gemma":[0.001272747,0.0004393143,0.0002886175,0.000125149,0.000103428,0.0000963927,0.00187181,0.0008966954,0.0000168018],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006693046,"about_ca_system_score_gemma":0.00009253796,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007741035,"about_ca_topic_score_gemma":0.00003838824,"domain_scores_codex":[0.9966118,0.00009609092,0.0007052603,0.001416785,0.0003708167,0.0007992242],"domain_scores_gemma":[0.9985282,0.0007333129,0.0002003322,0.0003257065,0.00004703818,0.0001653553],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007569481,0.0000910833,0.0002944516,0.0001043476,0.00004472085,8.328015e-7,0.000404804,0.9583068,0.004521808,0.000163686,0.0002478693,0.03574387],"study_design_scores_gemma":[0.0005523611,0.0001123451,0.00007900255,0.0000994462,0.00007490211,0.000009325842,0.000008948651,0.9844909,0.001772867,0.007060118,0.005302558,0.0004372592],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08445965,0.0000169541,0.9100076,0.0004093864,0.0002626109,0.00142908,0.00009194871,0.0004697409,0.002853016],"genre_scores_gemma":[0.1266906,0.000003694032,0.8680252,0.0004118004,0.00009196494,0.001054905,0.0003137839,0.00003676257,0.003371282],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04223091,"threshold_uncertainty_score":0.9998059,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0610815415135946,"score_gpt":0.3031462420867289,"score_spread":0.2420647005731343,"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."}}