{"id":"W4389584550","doi":"10.1109/tai.2023.3340982","title":"Manipulation Attacks on Learned Image Compression","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Artificial Intelligence","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; University of Waterloo","funders":"Army Research Office; Science and Technology Program of Hubei Province; National Natural Science Foundation of China","keywords":"Computer science; Image compression; Lossy compression; Lossless compression; Artificial intelligence; JPEG; Deep learning; Robustness (evolution); Image quality; Computer vision; Data compression; Computer engineering; Image processing; Image (mathematics)","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":[0.0004968513,0.0005259129,0.0003327207,0.0004464249,0.000214677,0.0004682064,0.0004754152,0.0005448815,0.001329877],"category_scores_gemma":[0.003233085,0.0001485046,0.0003373699,0.0002364134,0.0009763309,0.001075787,0.0009668169,0.000869928,0.0002225188],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000647273,"about_ca_system_score_gemma":0.0002998612,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001023226,"about_ca_topic_score_gemma":0.000657753,"domain_scores_codex":[0.9994431,0.0001115644,0.0000248473,0.00007508504,0.0002444893,0.0001007898],"domain_scores_gemma":[0.9986929,0.0006599102,0.0001749355,0.000318109,0.0001176739,0.00003650403],"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":[0.0005120718,0.00009778245,0.002466398,0.0001228272,0.00007521403,0.0003904329,0.0001145184,0.8235884,0.0709614,0.02557036,0.001590784,0.07450983],"study_design_scores_gemma":[0.000007229077,0.00008182104,0.0004492799,0.00001053442,0.000007354896,0.00008386057,0.000008866717,0.9642404,0.03070846,0.003748694,0.000644782,0.000008651371],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6613026,0.0008493774,0.322655,0.0008600143,0.00009556042,0.00009159935,0.0002651899,0.002469349,0.01141136],"genre_scores_gemma":[0.9866599,0.0001337864,0.01180661,0.00008233455,0.00001465102,0.00001923843,0.00008562313,0.00005405119,0.00114375],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.001329877,"threshold_uncertainty_score":0.004696369,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1014077847971648,"score_gpt":0.357783726830343,"score_spread":0.2563759420331782,"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."}}