{"id":"W4408353585","doi":"10.1109/icassp49660.2025.10888052","title":"INN-based Secure Steganography Using Lost Information as Adversarial Perturbations","year":2025,"lang":"en","type":"article","venue":"","topic":"Advanced Steganography and Watermarking Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Science Foundation of Guangdong Province","keywords":"Adversarial system; Steganography; Computer science; Computer security; Steganography tools; Artificial intelligence; Embedding","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":[0.0004676326,0.0004038423,0.0004205696,0.0003040574,0.0001621005,0.0002629223,0.0004487838,0.0003435619,0.0006653113],"category_scores_gemma":[0.0009624628,0.0001749991,0.0003023008,0.0001694403,0.0006088617,0.000832507,0.00073361,0.0005091528,0.0001806427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003806941,"about_ca_system_score_gemma":0.0002494322,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006402642,"about_ca_topic_score_gemma":0.0007215257,"domain_scores_codex":[0.9996133,0.00007239713,0.00001742127,0.00007414833,0.0001717168,0.00005105515],"domain_scores_gemma":[0.9995533,0.00019003,0.00007818998,0.00008810936,0.00007157231,0.00001878115],"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":[0.0007030942,0.0001264277,0.002397213,0.0002341751,0.0001248157,0.0005736132,0.0002262105,0.4274652,0.3342614,0.02630063,0.001072916,0.2065143],"study_design_scores_gemma":[0.000007676543,0.00008548434,0.0004206798,0.000009226515,0.00001260411,0.0002139277,0.00001240226,0.9264879,0.06932993,0.002513886,0.0008933045,0.00001304278],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.157179,0.0003821655,0.8373172,0.0001826745,0.00005831195,0.00004136261,0.00004278775,0.0008538915,0.003942541],"genre_scores_gemma":[0.9405591,0.0002005648,0.05652929,0.00005824789,0.00001570046,0.00002008391,0.00005141638,0.00003665621,0.002528983],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0006653113,"threshold_uncertainty_score":0.002762139,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008496895162379677,"score_gpt":0.2544651981435471,"score_spread":0.2459683029811674,"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."}}