{"id":"W4404660366","doi":"10.2139/ssrn.5031810","title":"Evaluating and Enhancing Segmentation Model Robustness with Metamorphic Testing","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Robustness (evolution); Segmentation; Computer science; Artificial intelligence; Robustness testing; Computer vision; Biology","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":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.004708724,0.0002757809,0.000251976,0.0002483299,0.0003230775,0.000879007,0.0006025718,0.0001144809,0.000001126919],"category_scores_gemma":[0.0002089549,0.0002224272,0.00005001377,0.0002976605,0.00002776709,0.0003268329,0.0007126365,0.004925094,0.00000494401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006310992,"about_ca_system_score_gemma":0.004318886,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005952056,"about_ca_topic_score_gemma":0.0002045678,"domain_scores_codex":[0.9970763,0.0002152999,0.0003930129,0.0006694839,0.0005312077,0.001114637],"domain_scores_gemma":[0.9987329,0.0001255683,0.0004500246,0.0004122837,0.0001990227,0.00008015193],"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":[0.00001118426,0.00001982796,0.0001842948,0.000161955,0.0001713161,0.000005747592,0.0005270425,0.6626183,0.003040582,0.05482071,0.000004990224,0.2784341],"study_design_scores_gemma":[0.000175611,0.0001422711,0.00005904902,0.0002658214,0.0001067985,0.000506231,0.0001610024,0.8442862,0.0000895217,0.1539836,0.000001654023,0.0002222028],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1413037,0.003002658,0.8541294,0.0009404013,0.0001551306,0.0001621546,0.000001503975,0.0001495041,0.0001555456],"genre_scores_gemma":[0.7961366,0.0003991366,0.2028079,0.0000316628,0.0001914873,0.00002677678,0.00001697955,0.0000332029,0.0003562968],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6548328,"threshold_uncertainty_score":0.9973706,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05007778346795639,"score_gpt":0.3265554576501277,"score_spread":0.2764776741821713,"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."}}