{"id":"W2954460650","doi":"","title":"Empirical Analysis of Beam Search Performance Degradation in Neural Sequence Models","year":2019,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":42,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Degradation (telecommunications); Sequence (biology); Computer science; Artificial neural network; Artificial intelligence; Pattern recognition (psychology); Chemistry; Telecommunications","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":[0.008425904,0.0006433566,0.0008631665,0.001056818,0.0004637208,0.0007937511,0.0009388309,0.001689385,0.001918339],"category_scores_gemma":[0.05754972,0.0004053525,0.0004291516,0.001207955,0.0007568896,0.001852863,0.0006860707,0.001414346,0.0006207352],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001274676,"about_ca_system_score_gemma":0.001147602,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008301697,"about_ca_topic_score_gemma":0.004825152,"domain_scores_codex":[0.9974291,0.001277781,0.0001772606,0.0003683934,0.0005253282,0.0002220872],"domain_scores_gemma":[0.9321679,0.0577975,0.001894883,0.003477601,0.004144928,0.0005171674],"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":[0.00110023,0.0002311242,0.01302465,0.0002932325,0.0001858905,0.0001192672,0.0001474031,0.90577,0.005094876,0.005649657,0.00378041,0.06460322],"study_design_scores_gemma":[0.00001671659,0.000143978,0.003719477,0.00002395603,0.00002279822,0.0001053619,0.00003239718,0.9917935,0.002503839,0.001332222,0.0002894466,0.00001638287],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7383288,0.005887026,0.2481458,0.001236709,0.0001455647,0.00008107369,0.00137411,0.001723573,0.00307726],"genre_scores_gemma":[0.9736143,0.0005160783,0.02245163,0.0001293645,0.00004109454,0.00004964319,0.001939078,0.0001605515,0.001098293],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008425904,"threshold_uncertainty_score":0.04456097,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1093036857200191,"score_gpt":0.3857356482147963,"score_spread":0.2764319624947772,"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."}}