{"id":"W2146546639","doi":"","title":"Learning Bigrams from Unigrams","year":2008,"lang":"en","type":"article","venue":"Meeting of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Bigram; Perplexity; Computer science; Language model; Natural language processing; Artificial intelligence; Word (group theory); Trigram; Oracle; Set (abstract data type); Speech recognition; Mathematics; Programming language","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":[0.001782699,0.001268043,0.001078867,0.003335209,0.000702576,0.001367571,0.001567557,0.00139422,0.00308919],"category_scores_gemma":[0.01007375,0.0007294053,0.001037871,0.001987878,0.0006651078,0.004932363,0.001828956,0.002323459,0.002976471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005470841,"about_ca_system_score_gemma":0.0009781776,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001620828,"about_ca_topic_score_gemma":0.002692252,"domain_scores_codex":[0.9990152,0.0003846548,0.00007620572,0.0002700074,0.0001419139,0.0001119058],"domain_scores_gemma":[0.9955534,0.002937123,0.0002739089,0.0006914452,0.0004186366,0.0001255359],"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":[0.000644828,0.0002719231,0.005878229,0.0005095004,0.0002304733,0.0006778223,0.0009754828,0.06455686,0.01448223,0.02929886,0.01020996,0.8722639],"study_design_scores_gemma":[0.00005026698,0.0001642416,0.001247036,0.00008242302,0.00006265422,0.0003087123,0.0003400806,0.8785776,0.008991661,0.1054612,0.004651956,0.00006213361],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1088765,0.001167862,0.881627,0.0004976408,0.0001578854,0.0000942168,0.0007399554,0.004354657,0.002484161],"genre_scores_gemma":[0.6478997,0.0009393764,0.339471,0.0003178644,0.000230418,0.0002246472,0.003431472,0.0006826097,0.006802874],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003335209,"threshold_uncertainty_score":0.01033437,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02239389677370867,"score_gpt":0.2445550466316511,"score_spread":0.2221611498579424,"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."}}