{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003900907,0.00008199878,0.0001256241,0.0001048036,0.000146675,0.0001120408,0.0003830796,0.00003978832,0.000002939354],"category_scores_gemma":[0.00004912133,0.00005666532,0.00004533043,0.0005089865,0.00003351226,0.0005624384,0.00009992408,0.0003665324,4.643143e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003526118,"about_ca_system_score_gemma":0.00001548427,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003528386,"about_ca_topic_score_gemma":0.000008228258,"domain_scores_codex":[0.999121,0.00009329805,0.0003495536,0.0001445742,0.0001467382,0.0001448741],"domain_scores_gemma":[0.9990033,0.000496483,0.0002439123,0.0001675678,0.0000530353,0.00003571851],"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":[0.00003153604,0.00007824088,0.001103194,0.000006880881,0.000009636628,0.00001347334,0.0002455846,0.01246156,0.002632556,0.0007805998,0.006941862,0.9756949],"study_design_scores_gemma":[0.0007672465,0.0002524464,0.066356,0.0002617989,0.00001619298,0.00007580379,0.00005246544,0.9122109,0.000304342,0.01555265,0.003993611,0.0001565369],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4799624,0.002701766,0.5070605,0.008985634,0.0006096916,0.0002756877,0.000001017119,0.00002564905,0.0003776248],"genre_scores_gemma":[0.9951921,0.0004557561,0.003629328,0.0005887376,0.0001099495,0.000001584536,0.000001116233,0.000003755224,0.00001765535],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9755383,"threshold_uncertainty_score":0.2310745,"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."}}