{"id":"W3174361912","doi":"10.1016/j.cose.2021.102371","title":"I-MAD: Interpretable malware detector using Galaxy Transformer","year":2021,"lang":"en","type":"article","venue":"Computers & Security","topic":"Advanced Malware Detection Techniques","field":"Computer Science","cited_by":43,"is_retracted":false,"has_abstract":false,"ca_institutions":"Defence Research and Development Canada; Queen's University; McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Defence Research and Development Canada; Nvidia","keywords":"Malware; Computer science; Interpretability; Artificial intelligence; Executable; Machine learning; Data mining; Detector; Pattern recognition (psychology); Computer security; Programming language","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001813927,0.0002904363,0.0003416226,0.0001508651,0.0002200144,0.0002484766,0.0009058147,0.0001342746,0.00006657361],"category_scores_gemma":[0.00004105364,0.0003198175,0.0002030606,0.000746636,0.00007014981,0.001108082,0.0003728666,0.0003826949,0.00002658777],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001955566,"about_ca_system_score_gemma":0.000152719,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003294486,"about_ca_topic_score_gemma":0.00002928097,"domain_scores_codex":[0.9978877,0.0001279112,0.0003702215,0.0007732494,0.0003242948,0.0005166252],"domain_scores_gemma":[0.9984428,0.0001011185,0.0001110487,0.0009020406,0.0002552445,0.0001877855],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001462951,0.001010582,0.001657583,0.0009217709,0.0004872384,0.002251362,0.01591453,0.00442491,0.1962489,0.04157808,0.01023906,0.7251197],"study_design_scores_gemma":[0.0008699283,0.0002379701,0.0005818984,0.0003048104,0.00003467952,0.0009423508,0.00009924999,0.2867857,0.6218852,0.04353594,0.04341492,0.001307392],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02656324,0.0003972536,0.9696543,0.0002705847,0.001176854,0.0002026683,0.000007935543,0.001138867,0.000588359],"genre_scores_gemma":[0.7014334,0.00002772242,0.2978431,0.0005194707,0.0000927709,0.00001326544,0.000003734331,0.0000228605,0.00004367813],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7238123,"threshold_uncertainty_score":0.9999254,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01242615308146986,"score_gpt":0.2565589677412584,"score_spread":0.2441328146597885,"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."}}