{"id":"W6891801165","doi":"10.48448/sy3s-4383","title":"A Multilingual Evaluation of NER Robustness to Adversarial Inputs","year":2023,"lang":"en","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Adversarial system; Robustness (evolution); Training set; German; Named-entity recognition; Focus (optics); Language model","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":[0.00634891,0.002160515,0.001040748,0.001107907,0.0008179585,0.001252825,0.001156854,0.001301428,0.004976561],"category_scores_gemma":[0.01283662,0.0003281422,0.0009689623,0.0007350154,0.001186527,0.002004732,0.00299684,0.001954417,0.002482959],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009810878,"about_ca_system_score_gemma":0.0006795666,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005239453,"about_ca_topic_score_gemma":0.007225861,"domain_scores_codex":[0.9951862,0.002003883,0.0002941799,0.00116151,0.001020398,0.0003338694],"domain_scores_gemma":[0.9922131,0.004167055,0.0003706167,0.001722651,0.001192327,0.000334405],"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":[0.00206586,0.001003174,0.01157793,0.001028324,0.00117046,0.0008497315,0.0004202877,0.6637313,0.04828224,0.005820844,0.02532742,0.2387224],"study_design_scores_gemma":[0.0001044267,0.001548548,0.008039895,0.0001293939,0.000195176,0.0007656563,0.0002701439,0.8822882,0.08704695,0.004936573,0.01450641,0.0001686696],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.642035,0.004522051,0.2834883,0.001407683,0.001388931,0.000578083,0.006243032,0.01437886,0.04595812],"genre_scores_gemma":[0.9172206,0.0005366531,0.05891702,0.0005116179,0.0001248334,0.0001757433,0.01041371,0.001139325,0.01096045],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00634891,"threshold_uncertainty_score":0.03357661,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1252091582445891,"score_gpt":0.4142154400405781,"score_spread":0.289006281795989,"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."}}