{"id":"W4415390017","doi":"10.1016/j.cag.2025.104458","title":"Evaluating graphical perception capabilities of Vision Transformers","year":2025,"lang":"en","type":"article","venue":"Computers & Graphics","topic":"Industrial Vision Systems and Defect Detection","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Universität Ulm","keywords":"Perception; Visualization; Visual perception; Convolutional neural network; Benchmark (surveying); Transformer; Human visual system model; Graphical model","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":[0.0004736698,0.000135524,0.0002322372,0.0004221602,0.00009311214,0.00002754111,0.00009192582,0.000183568,0.000009772264],"category_scores_gemma":[0.00002118656,0.0001317064,0.0001819326,0.0006841795,0.00008107242,0.0001062366,0.00001333432,0.0002242376,0.000002156097],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004273489,"about_ca_system_score_gemma":0.0000204291,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004172313,"about_ca_topic_score_gemma":0.00000960944,"domain_scores_codex":[0.9989765,0.00007193114,0.000395988,0.0001632099,0.0002281033,0.000164215],"domain_scores_gemma":[0.9995388,0.0001338088,0.00003424292,0.0001592429,0.00009544983,0.00003842446],"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":[0.000241807,0.0001698863,0.004136019,0.002119976,0.0004072421,0.000003974608,0.004577608,0.07003271,0.1363346,0.02003363,0.008946154,0.7529964],"study_design_scores_gemma":[0.002422821,0.0009217699,0.03166691,0.001177154,0.0001379505,0.00001249368,0.001451699,0.9439523,0.006068388,0.005151005,0.006387851,0.0006497049],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8889346,0.0001425088,0.1077319,0.00005057291,0.001799486,0.0002423054,0.000003982379,0.0002083038,0.0008863813],"genre_scores_gemma":[0.9991965,0.00005223726,0.0006077795,0.00003686867,0.00006564284,0.000009880329,0.000005017527,0.00001238174,0.0000136613],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8739195,"threshold_uncertainty_score":0.5370831,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02334548695330664,"score_gpt":0.3018931063226519,"score_spread":0.2785476193693453,"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."}}