{"id":"W4226024751","doi":"10.1017/s1351324922000134","title":"Real-world sentence boundary detection using multitask learning: A case study on French","year":2022,"lang":"en","type":"article","venue":"Natural Language Engineering","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Hanbat National University","keywords":"Computer science; Sentence; Punctuation; Natural language processing; Task (project management); Artificial intelligence; Boundary (topology); Multi-task learning; Speech recognition","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":[0.0004066936,0.0002600387,0.0002057901,0.0005206574,0.000575599,0.0002095353,0.0006308387,0.00004161056,0.00001284842],"category_scores_gemma":[0.0001444819,0.0002567102,0.00007756084,0.001181982,0.0000142642,0.0004326335,0.0005841809,0.0014494,0.000002883063],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005662536,"about_ca_system_score_gemma":0.00004625497,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002970733,"about_ca_topic_score_gemma":0.000197491,"domain_scores_codex":[0.9981845,0.0001263607,0.0002412976,0.0005334048,0.0005230192,0.0003913964],"domain_scores_gemma":[0.9991775,0.0001228389,0.0001098557,0.000456598,0.00005505776,0.00007812593],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006519222,0.0008036924,0.003119814,0.0002034887,0.0002374012,0.09134087,0.05109345,0.09150086,0.5480612,0.0009598776,0.00009105966,0.2125231],"study_design_scores_gemma":[0.0005565971,0.0004411627,0.0004304808,0.00005049872,0.00002632085,0.004651661,0.003857788,0.9672834,0.0216662,0.0000316329,0.0002641054,0.000740142],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.957678,0.001692521,0.03753209,0.00002649497,0.0008400078,0.000366013,0.000002785442,0.001829287,0.000032834],"genre_scores_gemma":[0.939446,9.030297e-7,0.06011215,0.00004276657,0.00010953,0.00004638908,0.000003489809,0.00003295596,0.0002058224],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8757825,"threshold_uncertainty_score":0.9999885,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01083990589589887,"score_gpt":0.2761478394611767,"score_spread":0.2653079335652778,"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."}}