{"id":"W4386566488","doi":"10.18653/v1/2023.findings-eacl.83","title":"Large Language Models are few(1)-shot Table Reasoners","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":80,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Vector Institute","funders":"","keywords":"Table (database); Context (archaeology); Computer science; Shot (pellet); Code (set theory); Natural language processing; Artificial intelligence; Programming language; Database; Geography","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":[0.003145807,0.001489161,0.0009781884,0.001002881,0.0006775719,0.002684024,0.002518933,0.001741386,0.01203944],"category_scores_gemma":[0.01647124,0.0008747914,0.002049199,0.0007336987,0.0009627004,0.006461349,0.002163971,0.003455807,0.005998746],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00109941,"about_ca_system_score_gemma":0.001634718,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004511195,"about_ca_topic_score_gemma":0.01415903,"domain_scores_codex":[0.9979501,0.0008602124,0.0001206038,0.0006307208,0.0003258155,0.00011242],"domain_scores_gemma":[0.9933094,0.004722429,0.0002314531,0.00117024,0.0003999748,0.0001664878],"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":[0.001232463,0.0004813036,0.005981897,0.001781869,0.0005441452,0.0005243173,0.001630128,0.1641717,0.01920535,0.06732272,0.07384965,0.6632746],"study_design_scores_gemma":[0.0001136874,0.0001384737,0.0005078824,0.0001119759,0.00009651441,0.0002058888,0.0003172718,0.875646,0.007850339,0.09568312,0.01927902,0.00004995974],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03216017,0.001109958,0.9250205,0.001466926,0.000272553,0.0002674234,0.002738631,0.03072747,0.006236422],"genre_scores_gemma":[0.2974021,0.0004442136,0.6852754,0.001238274,0.000147874,0.0003478703,0.00656905,0.001730536,0.006844722],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01203944,"threshold_uncertainty_score":0.04027599,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04316837089407304,"score_gpt":0.275202534745373,"score_spread":0.2320341638512999,"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."}}