{"id":"W4404781595","doi":"10.18653/v1/2024.findings-emnlp.564","title":"Machine Translation Hallucination Detection for Low and High Resource Languages using Large Language Models","year":2024,"lang":"en","type":"article","venue":"","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research","keywords":"Computer science; Machine translation; Artificial intelligence; Translation (biology); Resource (disambiguation); Natural language processing; Computer network","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003350294,0.00007196987,0.00006646504,0.00009877833,0.0001004607,0.0001783555,0.0000889879,0.00005668395,0.000004887348],"category_scores_gemma":[0.00001272517,0.000063459,0.00002759429,0.0001690256,0.000009402155,0.0006619816,0.00001862153,0.00005454061,0.000001661943],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003447013,"about_ca_system_score_gemma":0.00001230571,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001500517,"about_ca_topic_score_gemma":0.0001710763,"domain_scores_codex":[0.9993559,0.00003473747,0.0001225155,0.0002703509,0.0001022196,0.0001143203],"domain_scores_gemma":[0.9996545,0.0001033387,0.00002036758,0.0001700964,0.00002463492,0.00002706407],"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":[0.000009242615,0.00002889002,0.000005857848,0.0001677938,0.000009205865,0.000001430707,0.003551228,0.002270866,0.01888429,0.09060188,0.000007895183,0.8844614],"study_design_scores_gemma":[0.00014234,0.00002493546,0.0001172355,0.00001829674,0.000007963987,0.000007313301,0.00009316834,0.9867954,0.006883908,0.00532379,0.0005068663,0.00007876046],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03597916,0.001178924,0.9615496,0.0004266871,0.00005830576,0.000232362,0.00001362879,0.0002790416,0.0002822695],"genre_scores_gemma":[0.9559073,0.000006403754,0.04382961,0.00005327723,0.00004422918,0.0000124811,0.00002358505,0.000007306171,0.0001157644],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9845245,"threshold_uncertainty_score":0.2587783,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02065377648243963,"score_gpt":0.2803603852179046,"score_spread":0.2597066087354649,"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."}}