{"id":"W4413146879","doi":"10.1109/cvpr52734.2025.02186","title":"Preserve or Modify? Context-Aware Evaluation for Balancing Preservation and Modification in Text-Guided Image Editing","year":2025,"lang":"en","type":"article","venue":"","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"National Research Foundation","keywords":"Computer science; Context (archaeology); Image editing; Image (mathematics); Artificial intelligence; History","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":[],"consensus_categories":[],"category_scores_codex":[0.001407992,0.0001194651,0.0001502271,0.0002132358,0.0001321435,0.0002428629,0.0003741975,0.00008838605,0.00001310922],"category_scores_gemma":[0.0009340452,0.0001011958,0.00003154933,0.0005544255,0.00003193193,0.001587987,0.0001254978,0.00008463603,0.000001706943],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001342851,"about_ca_system_score_gemma":0.0001603715,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001275049,"about_ca_topic_score_gemma":0.0000707881,"domain_scores_codex":[0.9986009,0.0001109333,0.0004124364,0.000425945,0.0002594371,0.0001903289],"domain_scores_gemma":[0.9985718,0.0002721689,0.0001412338,0.0003635505,0.000618913,0.00003231113],"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":[0.0001228115,0.0002339309,0.004352681,0.0005881506,0.00003679786,0.000001328109,0.001296545,0.0001167822,0.05328191,0.2667918,0.01302826,0.660149],"study_design_scores_gemma":[0.0006833163,0.00003271222,0.008881238,0.0000832121,0.000008659323,9.802932e-7,0.0001537108,0.9132306,0.05862878,0.01752483,0.0006606269,0.0001113172],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006181664,0.00009422436,0.9854795,0.004688941,0.00009717544,0.001426323,0.000003921772,0.0002275351,0.00180074],"genre_scores_gemma":[0.9575776,0.00002447355,0.039736,0.0003132392,0.0000419997,0.0006487995,0.00003058339,0.000007153895,0.001620103],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.951396,"threshold_uncertainty_score":0.4126645,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07847100008251207,"score_gpt":0.3666936518639458,"score_spread":0.2882226517814337,"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."}}