{"id":"W4393156222","doi":"10.1609/aaai.v38i7.28571","title":"A New Benchmark and Model for Challenging Image Manipulation Detection","year":2024,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Image Processing Techniques and Applications","field":"Engineering","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Benchmark (surveying); Artificial intelligence; Computer science; Image (mathematics); Computer vision; Image manipulation; Pattern recognition (psychology); Cartography; 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":[0.001575211,0.002916144,0.001368715,0.002828274,0.0008141552,0.002062867,0.004573046,0.002815939,0.003176208],"category_scores_gemma":[0.005932175,0.0004875624,0.001203806,0.001827788,0.0008477249,0.002574432,0.0014374,0.002185003,0.001422446],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003402774,"about_ca_system_score_gemma":0.001454452,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0333231,"about_ca_topic_score_gemma":0.02964726,"domain_scores_codex":[0.9989153,0.0002140563,0.00005351771,0.0004323678,0.0002260849,0.0001586428],"domain_scores_gemma":[0.9981081,0.0006681705,0.0001985891,0.0002689648,0.0006226674,0.0001334445],"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":[0.0009326882,0.0007249181,0.007415972,0.0007531674,0.0002426739,0.0004480775,0.00007513112,0.7520175,0.005046461,0.007758819,0.07581216,0.1487723],"study_design_scores_gemma":[0.00002714657,0.00007071305,0.0008906941,0.00002793811,0.00002235398,0.00008820269,0.00002366863,0.9900417,0.00145137,0.003623273,0.003716083,0.00001691692],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3136598,0.01482512,0.5427781,0.006145351,0.002084175,0.001440084,0.06771922,0.02039692,0.03095124],"genre_scores_gemma":[0.6333212,0.002965138,0.2372924,0.001382263,0.0007232247,0.001356306,0.1077726,0.0008608498,0.01432604],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0333231,"threshold_uncertainty_score":0.06625831,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06288092941123159,"score_gpt":0.2993721743013264,"score_spread":0.2364912448900948,"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."}}