{"id":"W3214002337","doi":"","title":"Combiner: Full Attention Transformer with Sparse Computation Cost","year":2021,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Computation; Quadratic equation; Leverage (statistics); Transformer; Factorization; Autoregressive model; Theoretical computer science; Algorithm; Artificial intelligence; Mathematics; Voltage","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":[0.0007423826,0.001545343,0.0009687781,0.0007161662,0.0003598751,0.001090719,0.002252822,0.0008639214,0.01034595],"category_scores_gemma":[0.002607082,0.0006316253,0.001035841,0.0007420085,0.0006598488,0.003138267,0.002252714,0.001684629,0.003875363],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001026302,"about_ca_system_score_gemma":0.001438831,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007368817,"about_ca_topic_score_gemma":0.01326182,"domain_scores_codex":[0.9994584,0.00008741072,0.00002491961,0.0001794068,0.0001695139,0.00008036634],"domain_scores_gemma":[0.9994673,0.0002135096,0.00003888873,0.0001449905,0.00008659885,0.00004877801],"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":[0.0006081452,0.00020479,0.001071889,0.0003254866,0.0001308871,0.0003120731,0.0002386219,0.08173067,0.04097829,0.04232008,0.02423278,0.8078464],"study_design_scores_gemma":[0.00006693567,0.0001345168,0.0002737914,0.00001691363,0.00005387895,0.0001802698,0.00004069056,0.9386196,0.01550272,0.03876698,0.006320031,0.00002356674],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.009199589,0.0003464901,0.9771824,0.0001815439,0.0000740178,0.00009488253,0.0002540983,0.009788397,0.00287856],"genre_scores_gemma":[0.4503664,0.0007106242,0.5295938,0.0006889862,0.0001899712,0.0003964293,0.001580477,0.001335244,0.01513806],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01034595,"threshold_uncertainty_score":0.03461063,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06034433982791471,"score_gpt":0.1764075594391855,"score_spread":0.1160632196112708,"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."}}