{"id":"W2579534470","doi":"","title":"Training Data Enrichment for Infrequent Discourse Relations","year":2016,"lang":"en","type":"article","venue":"International Conference on Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Parsing; Computer science; Training set; Relation (database); Natural language processing; Artificial intelligence; Training (meteorology); Confidence interval; Quality (philosophy); Machine learning; Data mining; Statistics","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":[],"consensus_categories":[],"category_scores_codex":[0.0003140328,0.000156967,0.0001268935,0.0001673574,0.0001323152,0.0002158785,0.001916349,0.00005846129,0.00005007447],"category_scores_gemma":[0.003618608,0.0001182626,0.00004307343,0.0001025739,0.00007586392,0.0002430846,0.0003506139,0.00011141,0.00003923773],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001433127,"about_ca_system_score_gemma":0.000363407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004809281,"about_ca_topic_score_gemma":0.000003713283,"domain_scores_codex":[0.9983892,0.00002885416,0.0003447674,0.0005062629,0.0005349302,0.0001959704],"domain_scores_gemma":[0.9971665,0.0008977727,0.0002030694,0.0004674126,0.001185865,0.00007933894],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00001067716,0.00004730846,0.00004087094,0.000003022612,0.00002707231,0.000004970112,0.0001215189,0.0002766067,0.0000860484,0.9503368,0.001997226,0.04704785],"study_design_scores_gemma":[0.0003684966,0.00007320254,0.00007592183,0.0001439733,0.000007717819,0.000006006038,0.00001950437,0.2679378,0.0001301859,0.7131991,0.01782963,0.0002084601],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00006122432,0.00003579165,0.9840076,0.006962998,0.001666208,0.0001751194,0.0003456669,0.0002604256,0.006484964],"genre_scores_gemma":[0.5286862,0.00000419618,0.4701715,0.0002536436,0.0003647098,0.00002233084,0.000179769,0.000007794642,0.0003098719],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5286249,"threshold_uncertainty_score":0.482261,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1427233374709916,"score_gpt":0.4063764021825755,"score_spread":0.2636530647115839,"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."}}