{"id":"W4392669708","doi":"10.18653/v1/2023.ijcnlp-main.35","title":"Attacking Open-domain Question Answering by Injecting Misinformation","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Science Foundation","keywords":"Misinformation; Chen; Computer science; Domain (mathematical analysis); Question answering; Computational linguistics; Open domain; Natural language processing; Artificial intelligence; Linguistics; Library science; Philosophy; Mathematics; Geology; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01440672,0.001242756,0.00202216,0.00205825,0.002021971,0.003536411,0.00236368,0.005088559,0.003422288],"category_scores_gemma":[0.06888351,0.001040498,0.00130138,0.001723606,0.00271624,0.01180635,0.009269629,0.005738947,0.002719738],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001104088,"about_ca_system_score_gemma":0.001624068,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001998399,"about_ca_topic_score_gemma":0.001916265,"domain_scores_codex":[0.9822726,0.01047218,0.0007441504,0.001881249,0.003801056,0.0008288174],"domain_scores_gemma":[0.9203648,0.0592424,0.002287077,0.01272845,0.004350514,0.001026901],"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":[0.003468874,0.001427917,0.01403277,0.001209006,0.0009664129,0.002019268,0.00706737,0.06699548,0.04856407,0.1615744,0.1304536,0.5622209],"study_design_scores_gemma":[0.0001551993,0.0002154332,0.0008812359,0.0001251553,0.0002430018,0.000724368,0.001028114,0.711615,0.03316713,0.2234363,0.02831039,0.00009873464],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1114957,0.004233007,0.8292815,0.02019836,0.001437367,0.0005430674,0.001416251,0.01452057,0.01687424],"genre_scores_gemma":[0.8481891,0.000809074,0.1390285,0.004436037,0.0007291262,0.0002080329,0.001558838,0.0005507948,0.004490513],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01440672,"threshold_uncertainty_score":0.07619089,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03142707791162644,"score_gpt":0.3068473172510588,"score_spread":0.2754202393394324,"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."}}