{"id":"W2963905071","doi":"","title":"Twin Networks: Matching the Future for Sequence Generation","year":2018,"lang":"en","type":"article","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":43,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Polytechnique Montréal","funders":"","keywords":"Computer science; Inference; Task (project management); Sequence (biology); Forcing (mathematics); Generative grammar; Matching (statistics); Generative model; Artificial intelligence; Recurrent neural network; Simple (philosophy); Term (time); Algorithm; Machine learning; Artificial neural network; Mathematics","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":[0.0009687947,0.0002321655,0.0001922783,0.0001536135,0.00070832,0.0005617366,0.001131609,0.0001903053,0.00003040847],"category_scores_gemma":[0.00009317025,0.0001776254,0.000154495,0.000507057,0.00009198285,0.0005772688,0.0001784693,0.0002243274,0.00002888865],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001693297,"about_ca_system_score_gemma":0.0001237415,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005190957,"about_ca_topic_score_gemma":0.002001258,"domain_scores_codex":[0.9982491,0.0001475508,0.0003239184,0.0004615135,0.000267698,0.0005502088],"domain_scores_gemma":[0.9983649,0.0001772087,0.0001669939,0.0009239319,0.0002158031,0.000151135],"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":[0.00002949862,0.00008464847,0.0003817143,0.00001142481,0.00004460169,0.0000132393,0.0006418516,0.0007654757,0.01472593,0.1945551,0.02264935,0.7660971],"study_design_scores_gemma":[0.0001900115,0.00009339949,0.0007584458,0.00001850535,0.000012769,0.00009517612,0.00007595839,0.9550285,0.01579481,0.009484557,0.01817211,0.0002757348],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01082721,0.0004462961,0.9731962,0.01322682,0.0006540529,0.0006791028,0.00001190062,0.0006259696,0.0003324207],"genre_scores_gemma":[0.3822361,0.0001279558,0.6028478,0.0109105,0.002734325,0.0006595527,0.00001546929,0.00003621738,0.0004320773],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.954263,"threshold_uncertainty_score":0.7243353,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02315183810143337,"score_gpt":0.2514149227999537,"score_spread":0.2282630846985203,"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."}}