{"id":"W4416214887","doi":"10.1109/tmm.2025.3632696","title":"Fast and Effective Overwrite Attack Against DNN-Based Image Watermarking Models","year":2025,"lang":"","type":"article","venue":"IEEE Transactions on Multimedia","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"Fundamental Research Funds for the Central Universities; Natural Science Foundation of Chongqing; National Natural Science Foundation of China","keywords":"Digital watermarking; Robustness (evolution); Watermark; Noise (video); Image (mathematics); Watermarking attack; Vulnerability (computing); Artificial neural network","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":[0.001037025,0.001082406,0.000695597,0.0005129945,0.0002846642,0.0005450285,0.0008412151,0.0009432606,0.000795874],"category_scores_gemma":[0.004430626,0.0003120587,0.0007382407,0.0002439405,0.0008213756,0.002078553,0.00134273,0.001536565,0.0002468068],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006511984,"about_ca_system_score_gemma":0.0006007004,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001894055,"about_ca_topic_score_gemma":0.001889641,"domain_scores_codex":[0.9991766,0.0001622349,0.0000734678,0.0001690728,0.0002891808,0.0001294462],"domain_scores_gemma":[0.9985422,0.0006252689,0.0002473452,0.0002937648,0.0002306004,0.00006086619],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0006398557,0.0001292145,0.002424388,0.000143347,0.0001663676,0.0004133657,0.0001438706,0.6874285,0.07682173,0.02162114,0.001786517,0.2082817],"study_design_scores_gemma":[0.000004582971,0.00004371144,0.000095052,0.000003963047,0.000007485901,0.00004443622,0.000004052763,0.9877053,0.0107431,0.001116219,0.0002257977,0.000006286741],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2205867,0.0006357044,0.7727712,0.0003830053,0.0001519287,0.00008002586,0.0000617783,0.001680113,0.003649494],"genre_scores_gemma":[0.9234125,0.0002564631,0.0738292,0.0001259185,0.00003711029,0.00003719774,0.00007996659,0.00006654547,0.002155188],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001894055,"threshold_uncertainty_score":0.005484402,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01282977566655843,"score_gpt":0.2746773812058975,"score_spread":0.2618476055393391,"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."}}