{"id":"W4404781255","doi":"10.18653/v1/2024.findings-emnlp.790","title":"Topic Modeling: Contextual Token Embeddings Are All You Need","year":2024,"lang":"en","type":"article","venue":"","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Security token; Computer science; Computer security","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000553071,0.00006773922,0.0001161976,0.00008020676,0.0001531589,0.000195347,0.0001390694,0.00005015017,0.00121354],"category_scores_gemma":[0.0001120509,0.00005591514,0.0001176974,0.0002847722,0.00004379938,0.0001536826,0.00003380604,0.00008264705,0.0001461959],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004714635,"about_ca_system_score_gemma":0.00007710206,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002111411,"about_ca_topic_score_gemma":0.0008889993,"domain_scores_codex":[0.9991175,0.00009505456,0.0001468258,0.0001876352,0.000281784,0.00017118],"domain_scores_gemma":[0.9995782,0.0001768683,0.00001754118,0.00006216364,0.0000804727,0.00008470774],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00000879392,0.00004716805,0.001053366,0.00002097338,0.0002168532,0.00002845751,0.0258243,0.006588045,0.00009351343,0.8339137,0.01248115,0.1197237],"study_design_scores_gemma":[0.00009385777,0.00001133391,0.0001219451,0.00001996417,0.00005357872,8.153057e-7,0.01086956,0.6105509,0.00001491115,0.0499117,0.3281789,0.0001726027],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.2725784,0.001417255,0.2838048,0.03095469,0.00119173,0.0002218833,0.000005276424,0.0007375128,0.4090884],"genre_scores_gemma":[0.9526606,0.00002590771,0.003524471,0.0005977397,0.0004095069,0.00000444971,0.000004504808,0.000005450999,0.04276738],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.784002,"threshold_uncertainty_score":0.9996995,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09091584989938681,"score_gpt":0.4075066138841508,"score_spread":0.3165907639847639,"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."}}