{"id":"W2777929561","doi":"10.48550/arxiv.1712.08207","title":"Variational Attention for Sequence-to-Sequence Models","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Sequence (biology); Encoder; Computer science; Artificial neural network; Set (abstract data type); Gaussian; Algorithm; Latent variable; Random sequence; Artificial intelligence; Mechanism (biology); Mathematics; Physics","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":[0.0003817375,0.0002587952,0.0002660736,0.0002196346,0.0003521166,0.0002813914,0.00296679,0.0002594195,0.000007296422],"category_scores_gemma":[0.00006703468,0.0003217405,0.0002115769,0.0001455417,0.00005386219,0.000904939,0.001732559,0.0003032347,0.00005057148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003420995,"about_ca_system_score_gemma":0.0004207441,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002125863,"about_ca_topic_score_gemma":0.00002705867,"domain_scores_codex":[0.9978793,0.0000577246,0.0002092243,0.001354033,0.0001383793,0.0003613955],"domain_scores_gemma":[0.9972987,0.0000776905,0.0003103679,0.001781377,0.000352839,0.0001790859],"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":[0.000006525941,0.00001519633,0.0001161942,0.00002822183,0.00002188961,0.00002103856,0.00007176628,0.5262532,0.00008000245,0.4726322,0.00005405313,0.000699717],"study_design_scores_gemma":[0.0001834297,0.00001525615,0.0001276536,0.00006089376,0.00001907863,0.000001968123,0.000004450981,0.6413838,0.00001269136,0.3578427,0.0001215072,0.000226548],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03146081,0.00001064563,0.9648075,0.0004960613,0.0009614305,0.0005578189,0.00005626457,0.0001873111,0.001462138],"genre_scores_gemma":[0.9203211,0.00002215948,0.07718232,0.0001332179,0.000150911,0.000006976136,0.00003181898,0.00001406184,0.002137453],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8888603,"threshold_uncertainty_score":0.9999235,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2599454264256302,"score_gpt":0.242695120612733,"score_spread":0.01725030581289716,"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."}}