{"id":"W4415028296","doi":"10.1007/978-981-95-1999-6_12","title":"NormSoftmax Attention: Improving Transformer Model Performance","year":2025,"lang":"en","type":"book-chapter","venue":"Lecture notes in electrical engineering","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec","funders":"","keywords":"Adaptability; Transformer; Flexibility (engineering); Focus (optics); SIGNAL (programming language); Personalization","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.00007679524,0.0003682523,0.0003383016,0.0002936873,0.00007866613,0.00008020144,0.0006618978,0.0003523532,0.000007016788],"category_scores_gemma":[0.00002307171,0.0003506885,0.000152311,0.0003559782,0.00001326918,0.0001764481,0.00006575473,0.001115078,0.00001147189],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000167308,"about_ca_system_score_gemma":0.00009164733,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003250367,"about_ca_topic_score_gemma":0.000005007822,"domain_scores_codex":[0.9983996,0.000002955126,0.0003553131,0.0005496935,0.0002348389,0.0004576469],"domain_scores_gemma":[0.9992391,0.0001513627,0.00006158619,0.0004274421,0.0000443245,0.00007619976],"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":[0.000002977833,0.00001079443,0.000004723982,0.00008634428,0.00001572217,0.000005520124,0.00001478422,0.5262297,0.0009562648,0.07231693,0.00003840911,0.4003178],"study_design_scores_gemma":[0.0001232682,0.00003300232,0.00001746067,0.0001283239,0.00001408261,0.00000985017,1.026516e-8,0.9904456,0.0005239235,0.005128996,0.003207389,0.0003681357],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00007347461,0.0008205163,0.9884982,0.0004244741,0.0001374626,0.0002689795,0.00000235519,0.0002398563,0.009534692],"genre_scores_gemma":[0.840674,0.001185533,0.1258424,0.001186195,0.0007356465,0.0002171156,0.00004723456,0.0001661436,0.02994578],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8626558,"threshold_uncertainty_score":0.9998945,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005979224417218525,"score_gpt":0.1904811787473016,"score_spread":0.1845019543300831,"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."}}