{"id":"W4389519982","doi":"10.18653/v1/2023.findings-emnlp.97","title":"On the Risk of Misinformation Pollution with Large Language Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":60,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Ministry of Education - Singapore; Bộ Giáo dục và Ðào tạo; Ministry of Education, India; National Science Foundation","keywords":"Misinformation; Obstacle; Harm; Computer science; Internet privacy; Risk analysis (engineering); Computer security; Data science; Psychology; Political science; Business; Social psychology","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":[0.01830155,0.001254916,0.00118445,0.001886151,0.001261168,0.003589345,0.001526051,0.002565366,0.001560135],"category_scores_gemma":[0.1423586,0.0008005427,0.0009146158,0.001184273,0.002839861,0.007800676,0.004034901,0.003899651,0.0006290291],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001524094,"about_ca_system_score_gemma":0.001738106,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003063586,"about_ca_topic_score_gemma":0.003512913,"domain_scores_codex":[0.9834154,0.01173777,0.0004781472,0.001427647,0.00241589,0.0005250623],"domain_scores_gemma":[0.7144375,0.2475289,0.01053784,0.02013305,0.006085249,0.001277373],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002441773,0.0005887146,0.05333987,0.0008821363,0.0007040144,0.001454025,0.005398525,0.6387519,0.01479882,0.0769885,0.007834014,0.1968177],"study_design_scores_gemma":[0.00003557465,0.0002229397,0.001362902,0.00007114539,0.00009869508,0.0003707916,0.0003525866,0.9480388,0.008408533,0.03885371,0.002128588,0.00005578123],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3899303,0.00208804,0.5903274,0.006153452,0.0002277631,0.0002965846,0.0004145317,0.002787529,0.007774268],"genre_scores_gemma":[0.9352074,0.0003424722,0.06215362,0.0005770902,0.0001163236,0.00009335034,0.0002025248,0.0002212284,0.001086039],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01830155,"threshold_uncertainty_score":0.096789,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01591063612662517,"score_gpt":0.2270961714336405,"score_spread":0.2111855353070153,"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."}}