{"id":"W4396892464","doi":"10.1364/ofc.2024.m4k.2","title":"Pruning Attention in Transformers for Nonlinear Channel Compensation in Optical Systems","year":2024,"lang":"en","type":"article","venue":"","topic":"Optical Network Technologies","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Huawei Technologies (Canada)","funders":"","keywords":"Computer science; Nonlinear system; Compensation (psychology); Transformer; Electronic engineering; Channel (broadcasting); Control theory (sociology); Artificial intelligence; Electrical engineering; Telecommunications; Engineering; Voltage; Physics; Psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007205029,0.0003554847,0.0005367602,0.0004050398,0.0005784233,0.0008596227,0.0008936048,0.0005815114,0.002174517],"category_scores_gemma":[0.004555729,0.0001767138,0.0002637749,0.000444405,0.0007624981,0.001544776,0.0008421571,0.0007064665,0.0001847825],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005712874,"about_ca_system_score_gemma":0.0008057784,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002306711,"about_ca_topic_score_gemma":0.004193043,"domain_scores_codex":[0.9995875,0.0001168346,0.00001744696,0.00004337843,0.0001417276,0.00009319933],"domain_scores_gemma":[0.9978782,0.00156186,0.0001102562,0.0001716905,0.0002027607,0.00007528529],"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":[0.001295964,0.0003124413,0.002816376,0.0002865308,0.0001101406,0.0007546802,0.0005373216,0.35805,0.1152141,0.1447981,0.004139132,0.3716851],"study_design_scores_gemma":[0.00002857857,0.0001380483,0.0005788798,0.00001136813,0.00003150369,0.000157137,0.00004462537,0.9353529,0.0253159,0.03700463,0.001322539,0.00001387922],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3087543,0.0007208197,0.682713,0.0005928756,0.00009007299,0.00005619416,0.0000520813,0.000516235,0.006504407],"genre_scores_gemma":[0.9632792,0.0001961159,0.03453942,0.0001047068,0.00004015368,0.00001938569,0.00003261166,0.00004584304,0.001742413],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002306711,"threshold_uncertainty_score":0.007274508,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01529353225070271,"score_gpt":0.2370733111189081,"score_spread":0.2217797788682054,"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."}}