{"id":"W3133879190","doi":"10.48550/arxiv.2103.02190","title":"An Iterative Contextualization Algorithm with Second-Order Attention","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Contextualization; Computer science; Sentence; Value (mathematics); Embedding; Encoding (memory); Context (archaeology); Order (exchange); Algorithm; Artificial intelligence; Theoretical computer science; Natural language processing; Machine learning","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":[0.001252805,0.001635872,0.001344754,0.001988519,0.0008852033,0.001254318,0.002074342,0.00135888,0.006631366],"category_scores_gemma":[0.004438023,0.0007454977,0.001182319,0.001911532,0.0007560941,0.001991971,0.002561,0.001705395,0.002254845],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001026871,"about_ca_system_score_gemma":0.002030533,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006681126,"about_ca_topic_score_gemma":0.0130442,"domain_scores_codex":[0.9987602,0.0002999812,0.00006873043,0.0004628157,0.0002501251,0.0001581605],"domain_scores_gemma":[0.9987473,0.0005092092,0.00008218894,0.0002757061,0.0003213528,0.00006420683],"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":[0.0002468036,0.0001669772,0.001389979,0.0001777031,0.0001357204,0.0001218378,0.0004954059,0.04606727,0.01843865,0.02694323,0.01122122,0.8945953],"study_design_scores_gemma":[0.00007223376,0.0001459969,0.0008243661,0.00002832429,0.00008797956,0.0001515646,0.0001212691,0.9325041,0.01054206,0.04485124,0.010632,0.00003892634],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007672217,0.0002191814,0.9871783,0.0001841957,0.00006898768,0.0001351182,0.00006523331,0.003020712,0.001456047],"genre_scores_gemma":[0.1510368,0.0001851932,0.8414257,0.0002895181,0.0001666515,0.0004187166,0.0005591389,0.0005543577,0.005363988],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006681126,"threshold_uncertainty_score":0.02218413,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05022255834153722,"score_gpt":0.1947406401730657,"score_spread":0.1445180818315285,"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."}}