{"id":"W4406048761","doi":"10.1167/jov.25.1.4","title":"Benchmarking the speed–accuracy tradeoff in object recognition by humans and neural networks","year":2025,"lang":"en","type":"article","venue":"Journal of Vision","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institutes of Health; National Eye Institute; York University; Gordon and Betty Moore Foundation; Alfred P. Sloan Foundation","keywords":"Computer science; Benchmark (surveying); Artificial intelligence; Inference; Convolutional neural network; Task (project management); Artificial neural network; Object (grammar); Benchmarking; Categorization; Recurrent neural network; Machine learning; Pattern recognition (psychology); Computation; Cognitive neuroscience of visual object recognition; Speedup; Deep learning; Block (permutation group theory); Algorithm","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.004388747,0.001040019,0.0005582654,0.0009980595,0.0004306537,0.001316493,0.001094554,0.001485046,0.001864777],"category_scores_gemma":[0.0266651,0.0004041471,0.0006200522,0.0008106812,0.001203163,0.002953007,0.001708427,0.001443225,0.00117963],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009332529,"about_ca_system_score_gemma":0.0004649152,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01056389,"about_ca_topic_score_gemma":0.01088439,"domain_scores_codex":[0.9966781,0.001174246,0.0002440356,0.001145962,0.0005277502,0.0002298518],"domain_scores_gemma":[0.9873214,0.007198677,0.001024265,0.002660133,0.001392209,0.0004033642],"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":[0.006542354,0.00208095,0.186723,0.001373977,0.001428455,0.0004166737,0.002244283,0.2375958,0.03194955,0.008970435,0.03558057,0.485094],"study_design_scores_gemma":[0.000339412,0.003103926,0.2045992,0.0002326783,0.0003112123,0.000975084,0.001353279,0.7013234,0.04168304,0.02528354,0.02051689,0.0002784073],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9615578,0.002092728,0.02636359,0.0005169387,0.0001955938,0.0001143498,0.00297846,0.0008560516,0.005324625],"genre_scores_gemma":[0.9751372,0.0003147126,0.01560967,0.0001489166,0.00005739197,0.0001052799,0.006539269,0.0001607269,0.001926786],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01056389,"threshold_uncertainty_score":0.02321023,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01653382248718248,"score_gpt":0.2993192532275485,"score_spread":0.282785430740366,"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."}}