{"id":"W4395069151","doi":"10.1007/s11263-024-02069-9","title":"Position, Padding and Predictions: A Deeper Look at Position Information in CNNs","year":2024,"lang":"en","type":"article","venue":"International Journal of Computer Vision","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Guelph; York University; Toronto Metropolitan University; Vector Institute","funders":"","keywords":"Padding; Encoding (memory); Heuristics; Position (finance); Computer science; Convolutional neural network; Artificial intelligence; Boundary (topology); ENCODE; Filter (signal processing); Pattern recognition (psychology); Theoretical computer science; Algorithm; Computer vision; Mathematics; Computer security","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002230217,0.00009860473,0.0001027901,0.0005409502,0.00006788075,0.0003928546,0.000404568,0.00004683046,0.000008940514],"category_scores_gemma":[0.000011204,0.00008903548,0.00005910565,0.0002667058,0.00002615184,0.003638765,0.0002523328,0.0001885611,0.00002559222],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002829543,"about_ca_system_score_gemma":0.00004038401,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002552432,"about_ca_topic_score_gemma":0.000002539419,"domain_scores_codex":[0.9987685,0.00004071871,0.0005042896,0.0001487049,0.0004280635,0.0001097245],"domain_scores_gemma":[0.9991993,0.0001528606,0.0001718729,0.0001125109,0.0002976461,0.00006576351],"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":[0.0001745194,0.0002198214,0.001116644,0.00005109866,0.0001752592,0.0004358702,0.003223005,0.09627917,0.004626879,0.1400535,0.0140379,0.7396063],"study_design_scores_gemma":[0.0005636207,0.0002488335,0.01112328,0.0006181303,0.000009518123,0.00297116,0.00001120714,0.9442689,0.0004562532,0.02220882,0.01736944,0.0001508575],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05303664,0.0003097632,0.9400063,0.005177724,0.0012038,0.00009754556,0.000004928982,0.00005804744,0.0001052573],"genre_scores_gemma":[0.928387,0.0002347491,0.07026174,0.0006489663,0.0004227808,0.000006107973,0.00001545042,0.000006311962,0.00001688947],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8753504,"threshold_uncertainty_score":0.3788306,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0060274247605354,"score_gpt":0.2768037056435105,"score_spread":0.2707762808829751,"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."}}