{"id":"W4392083945","doi":"10.18653/v1/2024.emnlp-main.1033","title":"Enhanced Hallucination Detection in Neural Machine Translation through Simple Detector Aggregation","year":2024,"lang":"en","type":"article","venue":"","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Hallucinating; Detector; Computer science; Machine translation; Translation (biology); Software deployment; Artificial intelligence; Simple (philosophy); Machine learning; Rank (graph theory); Natural language processing; Telecommunications; Mathematics; Software engineering","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":[0.001813125,0.0009469588,0.0013592,0.0009926618,0.0005270909,0.001277264,0.0009548095,0.001195499,0.00166112],"category_scores_gemma":[0.005033061,0.0004189045,0.0007035345,0.001051168,0.0008043392,0.001970138,0.002429256,0.001175589,0.0009444525],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00054243,"about_ca_system_score_gemma":0.0007553671,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001423989,"about_ca_topic_score_gemma":0.002399504,"domain_scores_codex":[0.9989008,0.0003088873,0.0000743669,0.0002273881,0.0003595039,0.0001290215],"domain_scores_gemma":[0.9975547,0.0009815238,0.0001986358,0.0005523842,0.0005847358,0.0001280526],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0006989026,0.0003671314,0.003779371,0.0002302944,0.0002790557,0.000564451,0.0003871162,0.1518981,0.2383128,0.01314459,0.00438371,0.5859545],"study_design_scores_gemma":[0.00001649625,0.0001261209,0.001031219,0.000007966756,0.00003344821,0.0001822584,0.00003999411,0.9374257,0.05140139,0.007667202,0.002034922,0.00003324195],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06449138,0.0002840758,0.9311483,0.0002046943,0.00008495325,0.00006274924,0.00007352695,0.002074252,0.001576135],"genre_scores_gemma":[0.5886522,0.000178473,0.4069461,0.0002288889,0.0001021779,0.00009143299,0.00031193,0.0002138682,0.00327499],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.001813125,"threshold_uncertainty_score":0.009588838,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0093627393100804,"score_gpt":0.2806969907406583,"score_spread":0.2713342514305779,"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."}}