{"id":"W4386427483","doi":"10.1016/j.softx.2023.101508","title":"CoTranslate: A web-based tool for crowdsourcing high-quality sentence pair corpora","year":2023,"lang":"en","type":"article","venue":"SoftwareX","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Crowdsourcing; Computer science; Sentence; World Wide Web; Natural language processing; Quality (philosophy); Web application; Artificial intelligence","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":[0.01173662,0.00216662,0.001329988,0.007495178,0.002608276,0.002824854,0.002801837,0.001962108,0.04783018],"category_scores_gemma":[0.04283631,0.001281651,0.0011193,0.004523486,0.00136252,0.003641701,0.008786818,0.002519089,0.03411109],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001228886,"about_ca_system_score_gemma":0.004339381,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004303795,"about_ca_topic_score_gemma":0.01123605,"domain_scores_codex":[0.9911878,0.003718459,0.0007760084,0.001441562,0.002595446,0.0002806489],"domain_scores_gemma":[0.9688917,0.01369418,0.001500673,0.007634019,0.006898537,0.001380899],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001541706,0.0005130944,0.002347175,0.003821072,0.0004884371,0.001156586,0.004292174,0.007644097,0.05845457,0.01486725,0.5691193,0.3357545],"study_design_scores_gemma":[0.001050708,0.000466009,0.005199002,0.000631432,0.0001393036,0.0008792019,0.002378813,0.09342148,0.05266604,0.04786336,0.7947456,0.000559032],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.01740703,0.001158238,0.6840929,0.00163911,0.00197381,0.004597423,0.07722454,0.1717614,0.04014551],"genre_scores_gemma":[0.07664736,0.0005554031,0.7057592,0.001218025,0.0006975134,0.008915515,0.149322,0.02593403,0.03095112],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.04783018,"threshold_uncertainty_score":0.160008,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02795925307258903,"score_gpt":0.3009481217690872,"score_spread":0.2729888686964982,"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."}}