{"id":"W4404782142","doi":"10.18653/v1/2024.findings-emnlp.155","title":"MINERS: Multilingual Language Models as Semantic Retrievers","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research","keywords":"Computer science; Natural language processing; Artificial intelligence; Semantic computing; Linguistics; Semantic Web","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002371575,0.001813477,0.001086676,0.001770518,0.0004954616,0.002172251,0.002259021,0.001309826,0.005574932],"category_scores_gemma":[0.007845452,0.0006488059,0.001766806,0.001337575,0.0006379967,0.005639073,0.00334799,0.002324078,0.004629156],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005930889,"about_ca_system_score_gemma":0.001031513,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002379948,"about_ca_topic_score_gemma":0.005727854,"domain_scores_codex":[0.9982729,0.0007984776,0.000115605,0.0004341819,0.0002745212,0.000104304],"domain_scores_gemma":[0.9983408,0.0008780184,0.0001246961,0.000384904,0.0002083963,0.0000631001],"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":[0.001284345,0.0006715807,0.007278617,0.0009970382,0.0007897427,0.000506595,0.0007640971,0.1739297,0.01194511,0.04248201,0.03260173,0.7267495],"study_design_scores_gemma":[0.00006313453,0.0001832875,0.0006013774,0.00008257273,0.00009538422,0.0002359974,0.0002244288,0.9400545,0.004544083,0.04318873,0.01068255,0.00004393023],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1053097,0.00447321,0.8438929,0.001830717,0.000497449,0.0004630221,0.007445396,0.02509151,0.01099603],"genre_scores_gemma":[0.5069161,0.002094997,0.4587785,0.00100987,0.000290122,0.0006210421,0.01768345,0.001488204,0.01111767],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005574932,"threshold_uncertainty_score":0.01865005,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01441515127103011,"score_gpt":0.2992104052329831,"score_spread":0.284795253961953,"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."}}