{"id":"W4407241355","doi":"10.36227/techrxiv.173894988.81604432/v1","title":"Towards a Unified Evaluation Framework: Integrating Human Perception and Metrics for AI-Generated Images","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Perception; Computer science; Artificial intelligence; Data science; Computer vision; Psychology; Neuroscience","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":[0.0007707819,0.000291498,0.000298906,0.0003625715,0.0003564162,0.0004401628,0.0007696197,0.0003339979,0.0000159152],"category_scores_gemma":[0.0004732459,0.0002726326,0.00009196541,0.0008955211,0.00004735516,0.0002640335,0.001122769,0.0006342607,0.000001834393],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002242629,"about_ca_system_score_gemma":0.0002299945,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005857072,"about_ca_topic_score_gemma":0.00002362538,"domain_scores_codex":[0.99788,0.0001429906,0.0004293009,0.0009455968,0.0003480052,0.0002540686],"domain_scores_gemma":[0.9976503,0.000301953,0.0002330446,0.0008344901,0.0009082694,0.00007189176],"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":[0.000005642887,0.00005810425,0.0000496176,0.0001336892,0.00004721606,3.8804e-7,0.0005141423,0.01409806,0.003521666,0.3871143,0.002573531,0.5918837],"study_design_scores_gemma":[0.0001569306,0.00003169915,0.0004313043,0.00009254425,0.00004980545,0.000001070954,0.00004391725,0.7289413,0.001433113,0.2683066,0.0002634276,0.0002483795],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002493641,0.0002060323,0.9906773,0.002388128,0.0003391857,0.001803154,0.00002779209,0.0003630664,0.001701699],"genre_scores_gemma":[0.2737542,0.00008667512,0.723758,0.0006104647,0.0001429582,0.0009296333,0.0001714519,0.00001257371,0.0005340451],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7148432,"threshold_uncertainty_score":0.9999726,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07612113581608967,"score_gpt":0.3963464464994623,"score_spread":0.3202253106833726,"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."}}