{"id":"W4385573710","doi":"10.18653/v1/2022.emnlp-main.803","title":"SPE: Symmetrical Prompt Enhancement for Fact Probing","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Task (project management); Computer science; Object (grammar); Artificial intelligence; Subject (documents); Symmetry (geometry); Task analysis; Natural language processing; Machine learning; Pattern recognition (psychology); Mathematics; Engineering","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":[0.0002800341,0.00006084247,0.00007870389,0.00007998901,0.0001838562,0.0000597813,0.0005549269,0.0000102039,0.0001653976],"category_scores_gemma":[0.0000199517,0.00005516841,0.00004267355,0.0002553464,0.000003964942,0.0001227639,0.0004880696,0.0000776613,0.00001369344],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001029682,"about_ca_system_score_gemma":0.00005110522,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001143852,"about_ca_topic_score_gemma":5.459224e-7,"domain_scores_codex":[0.999069,0.00002159431,0.0001453015,0.0003013493,0.0002492308,0.0002135591],"domain_scores_gemma":[0.9995674,0.00005113916,0.00003383466,0.0002859114,0.00002206591,0.00003963551],"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":[0.00000832175,0.0002445066,0.0001535367,0.00003828403,0.00002435845,0.000006461156,0.001211778,0.004153131,0.002895629,0.5131328,0.006019237,0.472112],"study_design_scores_gemma":[0.00035753,0.0002088294,0.00004116981,0.000002933596,0.000002755108,0.000007731059,0.00005603004,0.9293023,0.01128285,0.007569405,0.05099133,0.0001770776],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005988278,0.00003032137,0.9879431,0.001660788,0.0003203707,0.0004381267,7.115715e-7,0.000109056,0.003509261],"genre_scores_gemma":[0.6872579,7.120765e-7,0.3085556,0.0004436178,0.00004500193,0.0002602767,0.000001396504,0.000004875877,0.00343063],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9251492,"threshold_uncertainty_score":0.2249703,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04301873187770343,"score_gpt":0.2730614404365639,"score_spread":0.2300427085588604,"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."}}