{"id":"W4283799331","doi":"10.1609/aaai.v36i10.21382","title":"Generation-Focused Table-Based Intermediate Pre-training for Free-Form Question Answering","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Question answering; Table (database); Leverage (statistics); Schema (genetic algorithms); Artificial intelligence; Natural language processing; Ask price; Language model; Task (project management); Information retrieval; Data mining","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":[0.00105445,0.000203409,0.0002302854,0.000163721,0.000533707,0.0002706178,0.002475053,0.00005451229,0.00004533892],"category_scores_gemma":[0.0005313586,0.0001812175,0.0001117573,0.0004931578,0.00009474369,0.0004560114,0.0005907369,0.0003070529,0.000003580169],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001321219,"about_ca_system_score_gemma":0.0002425733,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005440055,"about_ca_topic_score_gemma":0.00002339437,"domain_scores_codex":[0.9979004,0.00002454645,0.0005844095,0.0005543063,0.0005541561,0.0003822149],"domain_scores_gemma":[0.9985436,0.000108106,0.0004154068,0.0004195102,0.0004395185,0.00007381818],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009015082,0.00009454229,0.0001264128,0.00006081053,0.00001435475,3.406256e-7,0.003916002,0.01376116,0.0584935,0.7801355,0.0001544341,0.1431528],"study_design_scores_gemma":[0.00005894303,0.0002096356,0.00003380789,0.00007523986,0.000007738648,0.000001567565,0.000251503,0.7499764,0.1901292,0.05892136,0.0001630922,0.0001714814],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2218427,0.00001674918,0.7715905,0.003562541,0.001188503,0.0007955722,0.00002051626,0.000166777,0.0008162032],"genre_scores_gemma":[0.973558,0.000002964061,0.02554656,0.0003074058,0.0001575651,0.0002875801,0.000002324035,0.00001650359,0.0001210976],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7517153,"threshold_uncertainty_score":0.7389835,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1391244311080463,"score_gpt":0.3130940979136808,"score_spread":0.1739696668056346,"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."}}