{"id":"W3157230690","doi":"10.48550/arxiv.2105.00811","title":"CBench: Towards Better Evaluation of Question Answering Over Knowledge Graphs","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Computer science; Question answering; Benchmark (surveying); Benchmarking; Suite; Information retrieval; Vocabulary; Set (abstract data type); Graph; Syntax; Artificial intelligence; Natural language processing; Task (project management); Theoretical computer science; Programming language; Linguistics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009227139,0.0002284366,0.0002894991,0.0003118207,0.00006521398,0.00008411138,0.0009934797,0.0002556976,0.00005751721],"category_scores_gemma":[0.00005881068,0.0002840338,0.0002097842,0.0005148607,0.0000487758,0.0006229495,0.001459323,0.0003736182,0.000007448898],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002857933,"about_ca_system_score_gemma":0.0004830131,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003082586,"about_ca_topic_score_gemma":0.00008638547,"domain_scores_codex":[0.9979856,0.0003562246,0.0002505139,0.0009496537,0.000232494,0.0002255097],"domain_scores_gemma":[0.9978406,0.0000407206,0.0002377882,0.001192655,0.0006051854,0.00008303436],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001165027,0.0002087398,0.006380996,0.0003237674,0.0001981333,0.00007790311,0.002119568,0.6564914,0.0008566253,0.2943291,0.00006795572,0.03893418],"study_design_scores_gemma":[0.0003372684,0.00001559354,0.007714063,0.0002214319,0.0001120569,0.000001670562,0.00004658656,0.9499841,0.0008114212,0.0404485,0.00003606071,0.0002711894],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5695074,0.0002177604,0.4281753,0.00003267703,0.0006417388,0.0001492706,0.000001959588,0.00006891585,0.001205069],"genre_scores_gemma":[0.9945424,0.00008442231,0.005133224,0.00003325082,0.00006098256,0.000001332262,0.00001378844,0.00001126956,0.0001193198],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4250351,"threshold_uncertainty_score":0.9999612,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.127953138872214,"score_gpt":0.2417656969451876,"score_spread":0.1138125580729736,"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."}}