{"id":"W4385570025","doi":"10.18653/v1/2023.acl-long.385","title":"Few-shot In-context Learning on Knowledge Base Question Answering","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":57,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Knowledge base; Question answering; Executable; Schema (genetic algorithms); Artificial intelligence; Context (archaeology); Natural language; Natural language processing; Entity linking; Natural language understanding; Matching (statistics); Baseline (sea); Information retrieval; Programming language","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":[0.0005331436,0.00008009325,0.00008615173,0.0002039415,0.00006422246,0.00007009882,0.0002876259,0.00003875813,0.0000227476],"category_scores_gemma":[0.00007728191,0.00007707931,0.00002622287,0.000454727,0.000006588277,0.0002352016,0.000159346,0.0001786605,0.0006631163],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006134576,"about_ca_system_score_gemma":0.0000318664,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001008065,"about_ca_topic_score_gemma":0.0002188263,"domain_scores_codex":[0.9991161,0.00006981756,0.0001456184,0.0003068737,0.0001122286,0.0002493357],"domain_scores_gemma":[0.9995429,0.0001024556,0.0000215309,0.0002579517,0.00002302629,0.00005213676],"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":[0.000006769856,0.0000648552,0.009496639,0.00003691407,0.000005840253,0.00007866316,0.006478184,0.06480666,0.002897859,0.2508411,0.0004920011,0.6647945],"study_design_scores_gemma":[0.0001820044,0.00003943833,0.004388068,0.00006441899,5.028616e-7,0.000001950379,0.0001511465,0.9908943,0.001434513,0.0007358036,0.001988632,0.0001192644],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5237979,0.00004829844,0.4403269,0.0008706671,0.000577644,0.0001089108,9.0851e-8,0.0008685641,0.03340105],"genre_scores_gemma":[0.9926012,0.000006853619,0.003429709,0.000119976,0.0000568771,0.00000882211,8.595229e-7,0.000006745679,0.003768994],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9260876,"threshold_uncertainty_score":0.8523239,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05001398167547733,"score_gpt":0.3010416719042726,"score_spread":0.2510276902287952,"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."}}