{"id":"W4285138469","doi":"10.1162/tacl_a_00487","title":"Heterogeneous Supervised Topic Models","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Office of Naval Research; Simons Foundation; Alfred P. Sloan Foundation; National Science Foundation","keywords":"Computer science; Inference; Artificial intelligence; Outcome (game theory); Machine learning; Topic model; Bayes' theorem; Bayesian inference; Latent variable; Language model; Bayesian probability; Natural language processing; Probabilistic logic; Tone (literature); Linguistics","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.0008554866,0.00006227857,0.0001282668,0.00007776876,0.001573932,0.00002399307,0.00026322,0.00003367984,0.0001308687],"category_scores_gemma":[0.0006349026,0.00006355612,0.0002718547,0.0003929508,0.0000413357,0.0000259823,0.00001603247,0.0001095434,0.00000108819],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000380406,"about_ca_system_score_gemma":0.0002715308,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002249325,"about_ca_topic_score_gemma":0.00009579504,"domain_scores_codex":[0.9984648,0.0003088255,0.0002776756,0.0001204897,0.0006898881,0.0001383387],"domain_scores_gemma":[0.9976499,0.001203412,0.0002340324,0.00008528784,0.0007936546,0.00003370061],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000009354818,0.00007268327,0.0003046763,0.000004381412,0.00007489677,5.484512e-8,0.001082421,0.8695096,0.000001780148,0.1279194,0.0001497264,0.000870941],"study_design_scores_gemma":[0.0004479703,0.00004283109,0.0005759547,0.000003213326,0.000164935,4.421018e-7,0.0004929247,0.4539331,0.00002187395,0.4734356,0.07075175,0.0001294795],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007819326,0.00008126294,0.9752662,0.002335436,0.002604886,0.0005898018,0.0004812233,0.00008167722,0.01074018],"genre_scores_gemma":[0.980868,0.000002916399,0.01507467,0.0001275672,0.0001873446,0.00004549582,0.00003598215,0.000008103453,0.003649909],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9730487,"threshold_uncertainty_score":0.9997259,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04311022111236518,"score_gpt":0.3237508931693172,"score_spread":0.280640672056952,"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."}}