{"id":"W4416223010","doi":"10.48550/arxiv.2511.08985","title":"DeepTracer: Tracing Stolen Model via Deep Coupled Watermarks","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México; Indian Council of Medical Research; Institute for Catastrophic Loss Reduction","keywords":"Digital watermarking; Watermark; Task (project management); Key (lock); Reliability (semiconductor); Tracing","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003453067,0.00144179,0.001220784,0.001294838,0.0006253321,0.001895846,0.002740367,0.002276286,0.002749262],"category_scores_gemma":[0.01506692,0.0008964681,0.001624281,0.0008965433,0.002969733,0.007468327,0.005999027,0.002902966,0.0008330505],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001245651,"about_ca_system_score_gemma":0.001437555,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002179648,"about_ca_topic_score_gemma":0.002151486,"domain_scores_codex":[0.9977738,0.0005505843,0.0001445596,0.0005793533,0.0007006166,0.0002511383],"domain_scores_gemma":[0.9928744,0.002306221,0.001099481,0.003147495,0.0004043169,0.0001681589],"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":[0.0007914533,0.0002032692,0.004402692,0.0003189093,0.000247165,0.000516966,0.0003903793,0.6244731,0.02412546,0.07732038,0.003944558,0.2632657],"study_design_scores_gemma":[0.00001887599,0.00005722791,0.0001462636,0.00001799034,0.00001766966,0.0000638264,0.00001856503,0.9605477,0.008296096,0.0297997,0.001001462,0.0000146445],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03894841,0.0004200668,0.9557446,0.0003327178,0.00006032991,0.00007591766,0.0001863361,0.002973984,0.001257643],"genre_scores_gemma":[0.7753661,0.0004553343,0.2184277,0.0003338013,0.00007694905,0.0001318022,0.0007148478,0.0006073953,0.003885888],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003453067,"threshold_uncertainty_score":0.01826173,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02529531175863155,"score_gpt":0.2790813172186406,"score_spread":0.2537860054600091,"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."}}