{"id":"W4401201671","doi":"10.48550/arxiv.2407.19394","title":"Depth-Wise Convolutions in Vision Transformers for Efficient Training on Small Datasets","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Transformer; Training (meteorology); Computer science; Artificial intelligence; Computer vision; Pattern recognition (psychology); Engineering; Geography; Electrical engineering; Voltage","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002029602,0.0003025117,0.0002827172,0.0004081926,0.0001683729,0.00007751281,0.001129997,0.000267335,0.000003536769],"category_scores_gemma":[0.00002046862,0.0003474852,0.0001943411,0.000894152,0.00009886339,0.0001081531,0.000738671,0.0008836189,0.00005952211],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003254951,"about_ca_system_score_gemma":0.0001786375,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002203068,"about_ca_topic_score_gemma":0.000320798,"domain_scores_codex":[0.9978092,0.00005881692,0.000250194,0.001357362,0.00007705292,0.0004473786],"domain_scores_gemma":[0.9986006,0.0003046994,0.0001126368,0.0007729671,0.00004685563,0.0001622116],"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":[0.00002990194,0.0001024815,0.000009564847,0.00006329096,0.00002013604,0.0000584931,0.0002446367,0.7969744,0.00006427255,0.1960759,0.000358857,0.005998097],"study_design_scores_gemma":[0.0004880579,0.0000583944,0.0002096749,0.0002228217,0.00004170336,0.000002376242,0.00006944699,0.9549382,0.00008546127,0.04068393,0.002829812,0.0003701606],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06808616,0.00007020729,0.9281971,0.0006466731,0.0004523185,0.001082216,0.0002974009,0.000269694,0.0008981747],"genre_scores_gemma":[0.9945395,0.000084485,0.004740234,0.0001337435,0.0000456448,0.00002139412,0.0002178004,0.00002508961,0.0001921236],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9264533,"threshold_uncertainty_score":0.9998977,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1135221462489228,"score_gpt":0.2439901235274572,"score_spread":0.1304679772785344,"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."}}