{"id":"W2168801328","doi":"10.1162/089120103322711587","title":"Embedding Web-Based Statistical Translation Models in Cross-Language Information Retrieval","year":2003,"lang":"en","type":"article","venue":"Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":106,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Université du Québec à Montréal","funders":"","keywords":"Computer science; Cross-language information retrieval; Machine translation; Natural language processing; Information retrieval; Artificial intelligence; Process (computing); Translation (biology); Machine translation software usability; Language model; Embedding; Example-based machine translation; Programming language","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":[0.002948029,0.0006858851,0.0009962313,0.002062425,0.0006153194,0.002010587,0.0008486329,0.001592291,0.002892602],"category_scores_gemma":[0.01255923,0.0008248767,0.00107615,0.002539771,0.001093389,0.004334592,0.001423631,0.001257772,0.002479332],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001088434,"about_ca_system_score_gemma":0.001222439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007951599,"about_ca_topic_score_gemma":0.007478223,"domain_scores_codex":[0.9977093,0.001545737,0.0001244642,0.0002656126,0.0002706117,0.00008442893],"domain_scores_gemma":[0.9948868,0.003550359,0.0003219531,0.0006213525,0.0005588894,0.0000606088],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002724509,0.0002492967,0.001439004,0.0001870036,0.0001525784,0.0002326494,0.0003334114,0.6794626,0.003470942,0.06309647,0.00294401,0.2481596],"study_design_scores_gemma":[0.00001376532,0.00002664273,0.0001219701,0.000008505562,0.0000121795,0.00002934367,0.00002473704,0.9709681,0.0005532609,0.02736202,0.000867193,0.00001224516],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02968374,0.0006860976,0.9654546,0.0005011595,0.00005434126,0.00007006529,0.0001510215,0.001362248,0.002036771],"genre_scores_gemma":[0.5984632,0.00136372,0.3919006,0.0003444765,0.0001776127,0.000361001,0.00113453,0.0005012973,0.005753377],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007951599,"threshold_uncertainty_score":0.01581061,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01959166820152231,"score_gpt":0.3257819259314265,"score_spread":0.3061902577299042,"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."}}