{"id":"W3133566923","doi":"10.48550/arxiv.2103.03806","title":"MalBERT: Using Transformers for Cybersecurity and Malicious Software Detection","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Malware Detection Techniques","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Moncton","funders":"","keywords":"Malware; Computer science; Software; Transformer; Computer security; Android malware; Android (operating system); Encoder; Artificial intelligence; Machine learning; Software engineering; Operating system; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001624877,0.0003068518,0.0003204593,0.0002603233,0.0002650695,0.0001339199,0.0004729151,0.0003803071,0.000005172931],"category_scores_gemma":[0.00005557067,0.0004024336,0.0002139134,0.0004088031,0.00009361367,0.0006521548,0.0006033914,0.0004488924,0.000001019975],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003109754,"about_ca_system_score_gemma":0.0001230082,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003626381,"about_ca_topic_score_gemma":0.00048653,"domain_scores_codex":[0.9982172,0.00007534909,0.0001870007,0.001121221,0.00007123598,0.0003280046],"domain_scores_gemma":[0.9988383,0.0001069673,0.0001891725,0.000502886,0.0002235299,0.0001391387],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001169341,0.001162764,0.006748483,0.006648832,0.001932187,0.002657827,0.00918345,0.3839405,0.03573604,0.08600139,0.000158196,0.4646609],"study_design_scores_gemma":[0.001106229,0.0002452427,0.00051907,0.0002637279,0.0002261388,0.0001350503,0.0003920773,0.8821602,0.05280921,0.05973233,0.0009281302,0.001482521],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2212526,0.00007981418,0.7773554,0.00001878621,0.0003529576,0.0004501761,0.00001523601,0.0004302868,0.00004476798],"genre_scores_gemma":[0.9429768,0.000168284,0.05660572,0.00006585305,0.00003887656,0.000005068923,0.000007760547,0.00002610699,0.0001055246],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7217242,"threshold_uncertainty_score":0.9998428,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05626123619679819,"score_gpt":0.2013812306101907,"score_spread":0.1451199944133925,"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."}}