{"id":"W4412877234","doi":"10.1145/3711896.3736561","title":"A Survey on Trustworthy LLM Agents: Threats and Countermeasures","year":2025,"lang":"en","type":"article","venue":"","topic":"Advanced Malware Detection Techniques","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Trustworthiness; Computer security; Computer science; Internet privacy","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000272091,0.0001003347,0.0001102789,0.0001319856,0.00008725412,0.00009608336,0.0003028582,0.00004261292,0.00001113966],"category_scores_gemma":[0.00006745198,0.00008557509,0.00002130054,0.0003137544,0.00003492528,0.0002175585,0.000148038,0.00008310329,0.000009449487],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003585398,"about_ca_system_score_gemma":0.00002504629,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008625419,"about_ca_topic_score_gemma":0.0002247641,"domain_scores_codex":[0.9992311,0.00006175933,0.000117584,0.0003169669,0.0001349218,0.0001376451],"domain_scores_gemma":[0.9993271,0.0001277415,0.00003050215,0.0003983278,0.00007964807,0.00003672409],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.00004238738,0.0001330184,0.01693194,0.00001819457,0.00003609853,0.00001512125,0.0000960863,0.00003861643,0.0001457661,0.06623501,0.02585865,0.8904491],"study_design_scores_gemma":[0.001028394,0.0007242613,0.8003362,0.0001649663,0.000009824847,0.00001329681,0.00002146799,0.01581722,0.04275188,0.09709159,0.04133635,0.0007044994],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004911839,0.0001969171,0.9799963,0.0003351159,0.0002157116,0.0001616867,0.000003813498,0.0006184723,0.01356019],"genre_scores_gemma":[0.9825292,0.0001332786,0.0111478,0.002113623,0.000009644703,0.0000230814,0.000001532345,0.000005343454,0.004036497],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9776174,"threshold_uncertainty_score":0.3489651,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02918380244393913,"score_gpt":0.3131584008557762,"score_spread":0.283974598411837,"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."}}