{"id":"W2788463229","doi":"10.18653/v1/n18-1128","title":"Reusing Weights in Subword-Aware Neural Language Models","year":2018,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"Ministry of Education and Science of the Republic of Kazakhstan; Nvidia","keywords":"Reuse; Syllable; Morpheme; Computer science; Margin (machine learning); Embedding; Word (group theory); Layer (electronics); Aggregate (composite); Character (mathematics); Simple (philosophy); Artificial intelligence; Language model; Natural language processing; Speech recognition; Machine learning; Linguistics; Mathematics; Engineering","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":[0.001101959,0.001336156,0.001312256,0.001033921,0.00041159,0.001590765,0.002156971,0.001506687,0.002339967],"category_scores_gemma":[0.007059255,0.0009555357,0.0008061531,0.001305123,0.000469475,0.00443338,0.001673148,0.00263237,0.001681503],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005792335,"about_ca_system_score_gemma":0.0008280764,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005197744,"about_ca_topic_score_gemma":0.008413316,"domain_scores_codex":[0.9995165,0.0001578772,0.00004472621,0.000150666,0.00006999082,0.0000603328],"domain_scores_gemma":[0.9980046,0.001230872,0.00009330046,0.0002565461,0.0003387245,0.00007604482],"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":[0.0004463862,0.0002339192,0.001228149,0.0002367514,0.0003281017,0.0001155236,0.0002035161,0.4641516,0.008886361,0.01137967,0.005503633,0.5072864],"study_design_scores_gemma":[0.000008044327,0.00001563135,0.00005864444,0.000006945505,0.00003070667,0.00001134652,0.000008072016,0.9885484,0.0008695086,0.01015408,0.0002828706,0.0000057975],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0708045,0.00324746,0.9194655,0.0007138014,0.0003621701,0.00006406513,0.0003051151,0.002623937,0.002413317],"genre_scores_gemma":[0.8157458,0.001927226,0.172987,0.0003298281,0.0004059674,0.0001882356,0.001145005,0.0007250581,0.006545905],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005197744,"threshold_uncertainty_score":0.01033497,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04677271909570187,"score_gpt":0.2847590447914972,"score_spread":0.2379863256957954,"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."}}