{"id":"W4200633215","doi":"10.1609/aaai.v36i10.21293","title":"From Good to Best: Two-Stage Training for Cross-Lingual Machine Reading Comprehension","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Artificial intelligence; Set (abstract data type); Natural language processing; Reading (process); Comprehension; Task (project management); Resource (disambiguation); Reading comprehension; Machine learning; Recall; Precision and recall; Linguistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.008368791,0.003392127,0.002162018,0.002058301,0.001213638,0.002451464,0.005534811,0.004948859,0.006220314],"category_scores_gemma":[0.02096722,0.001574254,0.001783408,0.001241357,0.001371187,0.005462499,0.004925737,0.006330049,0.003291523],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001435456,"about_ca_system_score_gemma":0.001999498,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005241988,"about_ca_topic_score_gemma":0.008757806,"domain_scores_codex":[0.9960335,0.001368375,0.0002546047,0.00150172,0.0004567903,0.0003851183],"domain_scores_gemma":[0.9859529,0.009406245,0.0004158195,0.001673641,0.00200112,0.0005503042],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002592012,0.002084219,0.02124793,0.000613252,0.0005915003,0.0006876708,0.001262154,0.194021,0.01642412,0.004473211,0.02324506,0.7327577],"study_design_scores_gemma":[0.0000949644,0.000303944,0.001130334,0.00003565226,0.00006458161,0.00007576365,0.0001070764,0.9868129,0.004878275,0.005486899,0.0009710355,0.00003874991],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.311179,0.003246742,0.643713,0.002707863,0.0004395435,0.001013821,0.001595434,0.02667898,0.009425698],"genre_scores_gemma":[0.7401291,0.0002365417,0.2467435,0.001598548,0.0001715163,0.0007631623,0.004445238,0.0007561521,0.005156214],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008368791,"threshold_uncertainty_score":0.04425889,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.172198085721702,"score_gpt":0.3666931078557866,"score_spread":0.1944950221340846,"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."}}