{"id":"W2986265153","doi":"10.1162/coli_a_00367","title":"On the Linguistic Representational Power of Neural Machine Translation Models","year":2020,"lang":"en","type":"preprint","venue":"Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Computer science; Natural language processing; Machine translation; Artificial intelligence; Interpretability; Semantics (computer science); Rule-based machine translation; Word (group theory); Contrast (vision); Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003001909,0.0003100345,0.0003304252,0.0001545815,0.0001377982,0.0001842342,0.001512714,0.0001703825,0.00001398744],"category_scores_gemma":[0.004135843,0.000250916,0.0001763704,0.0002837665,0.0001195653,0.00003883559,0.0006349058,0.0008967537,0.00000622685],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005547177,"about_ca_system_score_gemma":0.0003201678,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002154437,"about_ca_topic_score_gemma":8.647553e-7,"domain_scores_codex":[0.9974905,0.0001463559,0.0006356502,0.0006228272,0.0009199234,0.0001847907],"domain_scores_gemma":[0.9957255,0.001815159,0.0005253474,0.0005789181,0.001275722,0.00007938723],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000018869,0.00004321612,0.000006423096,0.00007339352,0.00003149264,0.00001653547,0.0008325799,0.2820479,0.000005472172,0.7153385,0.0005710868,0.001014534],"study_design_scores_gemma":[0.00006608821,0.0000311585,0.000018494,0.00006640128,0.0000122864,0.000001933191,0.000001863029,0.507749,0.00005230776,0.4918212,0.00005604725,0.0001232411],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001439908,0.001924183,0.9909416,0.002125944,0.00124505,0.000388814,0.0001667696,0.0003468943,0.002716798],"genre_scores_gemma":[0.6159871,0.000003126723,0.3829848,0.0005540061,0.0002477619,0.00001313806,0.0001835144,0.0000182976,0.000008316805],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6158431,"threshold_uncertainty_score":0.9999943,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05564198337220849,"score_gpt":0.3233539129337652,"score_spread":0.2677119295615567,"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."}}