{"id":"W6947764557","doi":"10.48448/yhcb-wf82","title":"The Paradox of the Compositionality of Natural Language: A Neural Machine Translation Case Study","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Principle of compositionality; Artificial neural network; Machine translation; Natural (archaeology); Meaning (existential); Natural language; Translation (biology); Semantics (computer science)","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000548011,0.0001214366,0.0001373391,0.00003819752,0.0004332445,0.0000234361,0.0006749866,0.00002853055,0.02342944],"category_scores_gemma":[0.00002839318,0.00006945608,0.00006746893,0.0006629509,0.001394845,0.0000543504,0.0003077931,0.0001990282,0.00001096861],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000211828,"about_ca_system_score_gemma":0.00003039015,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004657682,"about_ca_topic_score_gemma":0.01666979,"domain_scores_codex":[0.998373,0.0001258384,0.0002342315,0.0002419839,0.0008517308,0.0001731944],"domain_scores_gemma":[0.9992021,0.00007386518,0.00025828,0.0004172005,0.00001313425,0.00003547006],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0005123922,0.01133082,0.3809063,0.0003905185,0.0005119369,0.001028559,0.03904962,0.002927312,0.03926681,0.0087787,0.2035972,0.3116998],"study_design_scores_gemma":[0.007452426,0.001668014,0.3359544,0.0001080496,0.000675586,0.002809885,0.2653587,0.1239619,0.001859346,0.000306193,0.2572794,0.002566122],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8321074,0.00275369,0.00002932799,0.001836131,0.001263097,0.002054261,0.001511652,0.00009187921,0.1583525],"genre_scores_gemma":[0.9949812,0.000008127485,0.00002626612,0.00007077921,0.00001612603,0.00001251982,0.0000451436,0.00001968348,0.004820201],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3091337,"threshold_uncertainty_score":0.9774633,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0228908506364822,"score_gpt":0.2973939379624443,"score_spread":0.2745030873259621,"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."}}