{"id":"W4400525527","doi":"10.1145/3626772.3657675","title":"Towards Robust QA Evaluation via Open LLMs","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Universitas Brawijaya","keywords":"Computer science","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02737895,0.002053199,0.001362758,0.004013799,0.001273,0.006869153,0.003631929,0.003398382,0.01206902],"category_scores_gemma":[0.1308921,0.001215626,0.001886265,0.002285756,0.002037263,0.009441689,0.01016515,0.004655229,0.007173553],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003102667,"about_ca_system_score_gemma":0.003717867,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005409557,"about_ca_topic_score_gemma":0.005950412,"domain_scores_codex":[0.9632707,0.02429296,0.002165613,0.003040403,0.006413519,0.000816857],"domain_scores_gemma":[0.9322445,0.03826269,0.002417326,0.01340741,0.01230187,0.001366207],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002174049,0.001240462,0.00995551,0.002390215,0.000494199,0.0004200044,0.003396235,0.1022098,0.030453,0.06678805,0.117091,0.6633875],"study_design_scores_gemma":[0.0002793253,0.0003952492,0.002095963,0.0003037775,0.0001004624,0.0002096436,0.0005693724,0.8459623,0.01871243,0.09152348,0.03971129,0.0001366941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04249916,0.001560434,0.8488861,0.001834276,0.0003745899,0.0009742986,0.00376507,0.0890638,0.01104227],"genre_scores_gemma":[0.4330677,0.0004242407,0.5383869,0.001305905,0.0002216669,0.001554387,0.01431758,0.006785038,0.00393644],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.972621,"threshold_uncertainty_score":0.1447954,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1242644351317595,"score_gpt":0.3507526136183011,"score_spread":0.2264881784865415,"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."}}