{"id":"W4408146237","doi":"10.1109/icmla61862.2024.00079","title":"Segmentation of Maya Hieroglyphs through Fine-Tuned Foundation Models","year":2024,"lang":"en","type":"article","venue":"","topic":"Image Processing and 3D Reconstruction","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mount Royal University","funders":"","keywords":"Maya; Foundation (evidence); Computer science; Segmentation; Artificial intelligence; Computer graphics (images); Archaeology; Geography","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008696445,0.0008378093,0.0004859554,0.002936962,0.0009547971,0.002923938,0.001024092,0.001259268,0.002134519],"category_scores_gemma":[0.002325367,0.0005013472,0.001165794,0.001751289,0.001080766,0.001306655,0.001679183,0.001434856,0.00282397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001223706,"about_ca_system_score_gemma":0.001674226,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01466907,"about_ca_topic_score_gemma":0.04575753,"domain_scores_codex":[0.999411,0.0001043597,0.00003197107,0.00024645,0.00009569077,0.0001106637],"domain_scores_gemma":[0.9992872,0.0002351717,0.00005311938,0.0002152178,0.0001493602,0.00005999025],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001028565,0.0003863892,0.02595434,0.0007014241,0.0002293353,0.001050772,0.003706873,0.1965169,0.04925399,0.02267662,0.06432439,0.6341704],"study_design_scores_gemma":[0.0000498847,0.00007346071,0.00994509,0.0001808113,0.00008418354,0.0005237141,0.002038398,0.8877778,0.02206744,0.01905322,0.05813638,0.00006969815],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3867291,0.001592946,0.5503955,0.001371772,0.0003191412,0.00044719,0.009201764,0.02044091,0.02950163],"genre_scores_gemma":[0.6632851,0.0005754606,0.3056011,0.0003183297,0.00008251866,0.0002483386,0.01954295,0.001807761,0.008538454],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01466907,"threshold_uncertainty_score":0.02916735,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02436218396478953,"score_gpt":0.2760455892150011,"score_spread":0.2516834052502116,"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."}}