{"id":"W4389520137","doi":"10.18653/v1/2023.findings-emnlp.377","title":"Disentangling Structure and Style: Political Bias Detection in News by Inducing Document Hierarchy","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea; Korea Advanced Institute of Science and Technology; Iran Telecommunication Research Center; National Research Foundation of Korea; National Research Foundation","keywords":"Computer science; Rhetorical question; Generalizability theory; Robustness (evolution); Style (visual arts); Overfitting; Writing style; Artificial intelligence; Journalism; Natural language processing; Sentence; Topic model; Information retrieval; Politics; Hierarchy; Linguistics; Sociology; Psychology; Political science; Literature; Art","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":[],"consensus_categories":[],"category_scores_codex":[0.000141872,0.00008379233,0.0000901833,0.0001492889,0.00005888805,0.0001443672,0.0001724659,0.00005240651,0.000008741286],"category_scores_gemma":[0.000044633,0.0000739512,0.00001448644,0.0003831359,0.00001434773,0.0002989617,0.000227565,0.0001499533,0.000004671702],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007598189,"about_ca_system_score_gemma":0.00001803653,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004493533,"about_ca_topic_score_gemma":0.0002266978,"domain_scores_codex":[0.9989684,0.00004636862,0.0001702987,0.0003245724,0.0001726363,0.0003176921],"domain_scores_gemma":[0.9995827,0.00007266357,0.00002012469,0.0002124904,0.00001060803,0.0001013894],"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":[0.00000958972,0.00003113197,0.05803518,0.00006977026,0.00002184931,0.00007416638,0.004524832,0.001375852,0.0176914,0.2327503,0.0004090267,0.6850069],"study_design_scores_gemma":[0.0006405472,0.00005242418,0.006252157,0.00003990379,0.00000370912,0.00002244615,0.0007506633,0.9176502,0.02649041,0.04717376,0.0005925993,0.0003311442],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8753077,0.00002296021,0.1222979,0.001714464,0.0001880474,0.00009067321,9.74252e-7,0.0001600207,0.0002172051],"genre_scores_gemma":[0.9938738,0.000005176931,0.005687177,0.0002605508,0.0000544257,0.000003928531,0.00000149239,0.000005297879,0.0001081584],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9162744,"threshold_uncertainty_score":0.3015642,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02283336190972059,"score_gpt":0.2674290424661955,"score_spread":0.2445956805564749,"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."}}