{"id":"W6929207923","doi":"10.48448/h1z6-se91","title":"A State-Vector Framework For Dataset Effects","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Feature (linguistics); Identification (biology); Context (archaeology); Field (mathematics); Data collection","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001656998,0.0005786374,0.0005891478,0.001379283,0.0004769137,0.000285013,0.002683675,0.0002169247,0.01040213],"category_scores_gemma":[0.002011998,0.000543933,0.00009798525,0.002418725,0.001410075,0.0002335583,0.0009363837,0.0007375327,0.003891655],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006857705,"about_ca_system_score_gemma":0.001421893,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000344602,"about_ca_topic_score_gemma":0.000524528,"domain_scores_codex":[0.9950238,0.0001216551,0.0003554259,0.001587028,0.001681932,0.001230089],"domain_scores_gemma":[0.9963943,0.0007654665,0.0005224211,0.001854645,0.00008491394,0.0003782585],"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":[0.00002733793,0.0001515075,0.00002884508,0.0001722633,0.00004365629,0.00002304864,0.00006324859,0.00003536886,0.0004616293,0.01131773,0.9854826,0.002192775],"study_design_scores_gemma":[0.0005375332,0.000304165,0.00002300981,0.0001506875,0.00006984921,0.000008767688,0.00004856432,0.001642838,0.0001876148,0.009030993,0.9872855,0.0007105101],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.0001913481,0.003789895,0.1415835,0.001262249,0.01531797,0.01936767,0.6080315,0.006507537,0.2039484],"genre_scores_gemma":[0.001860575,0.00008541525,0.5038523,0.004234233,0.00300583,0.002535297,0.07072386,0.007708937,0.4059936],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.5373076,"threshold_uncertainty_score":0.9997012,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02369200034604425,"score_gpt":0.3383174854454605,"score_spread":0.3146254850994163,"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."}}