{"id":"W4409327148","doi":"10.1109/access.2025.3559794","title":"Vision Transformers Versus Convolutional Neural Networks: Comparing Robustness by Exploiting Varying Local Features","year":2025,"lang":"en","type":"article","venue":"IEEE Access","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Robustness (evolution); Convolutional neural network; Artificial intelligence; Transformer; Pattern recognition (psychology); Computer vision; Voltage; Engineering; Electrical engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003497245,0.001974282,0.0009013055,0.001856058,0.0003917063,0.001249359,0.001109903,0.001305426,0.001344966],"category_scores_gemma":[0.0108681,0.0002883474,0.001105569,0.001046676,0.0008194438,0.00294732,0.001126955,0.001345601,0.0006147854],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001006219,"about_ca_system_score_gemma":0.0006997919,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005855332,"about_ca_topic_score_gemma":0.004978933,"domain_scores_codex":[0.9980793,0.0003767789,0.0002210681,0.000440306,0.0006591048,0.0002233685],"domain_scores_gemma":[0.9970733,0.001184462,0.0003701175,0.0006671755,0.0005677855,0.0001370922],"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":[0.004175098,0.0007285372,0.02147229,0.001584726,0.001514725,0.0004380254,0.000164159,0.462869,0.03430124,0.003495571,0.01123658,0.45802],"study_design_scores_gemma":[0.0001401831,0.003275794,0.01752198,0.000205719,0.0007361764,0.000531946,0.0002026765,0.9076538,0.05903384,0.00534652,0.005219616,0.0001319144],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8015877,0.01762896,0.1570677,0.001641439,0.001229279,0.0004761096,0.0038528,0.006435859,0.01008021],"genre_scores_gemma":[0.9551555,0.002585047,0.0340318,0.000325723,0.0001485053,0.0001414547,0.004883125,0.0003170649,0.002411837],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005855332,"threshold_uncertainty_score":0.01849538,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02814794017319139,"score_gpt":0.3123821569827348,"score_spread":0.2842342168095434,"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."}}