{"id":"W4412873526","doi":"10.2139/ssrn.5358561","title":"Dynamic Mutual Adversarial Learning for Semi-Supervised Semantic Segmentation of Underwater Images with Limited and Noisy Annotations","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Image Enhancement Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Carleton University","funders":"","keywords":"Segmentation; Adversarial system; Computer science; Underwater; Artificial intelligence; Computer vision; Pattern recognition (psychology); Machine learning; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009493137,0.0002838601,0.0003427038,0.0004213935,0.0002403238,0.000215978,0.0005819696,0.0001428079,0.000002510594],"category_scores_gemma":[0.00005384476,0.0002557998,0.000108612,0.0002251282,0.00007420611,0.0004610311,0.0003474352,0.001680399,6.504993e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005211686,"about_ca_system_score_gemma":0.002053905,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000378263,"about_ca_topic_score_gemma":0.0001651049,"domain_scores_codex":[0.9976122,0.0001558593,0.000441278,0.0004587939,0.0003187951,0.001013038],"domain_scores_gemma":[0.9987534,0.0001408845,0.0004393629,0.0002752135,0.0003478822,0.00004324843],"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":[0.002487363,0.001437919,0.007301886,0.004149225,0.008249712,0.00005595405,0.02275789,0.06397374,0.2041714,0.09231432,0.0009241824,0.5921764],"study_design_scores_gemma":[0.006438403,0.003320709,0.0007239726,0.001487669,0.0007781265,0.0003043227,0.002884924,0.7109889,0.04963585,0.2218239,0.0001591956,0.00145406],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03806972,0.0004952317,0.9597169,0.0006588247,0.0001727103,0.0006712138,0.000009083655,0.0001295067,0.00007684624],"genre_scores_gemma":[0.8923578,0.001640662,0.1044695,0.00005181098,0.00006030505,0.00008146107,0.00007813913,0.00002553393,0.001234808],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8552473,"threshold_uncertainty_score":0.9999894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007017174008908051,"score_gpt":0.2579206903466361,"score_spread":0.2509035163377281,"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."}}