{"id":"W4402670684","doi":"10.18653/v1/2024.findings-acl.151","title":"Accurate and Nuanced Open-QA Evaluation Through Textual Entailment","year":2024,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Textual entailment; Computer science; Natural language processing; Logical consequence; Artificial intelligence; Information retrieval","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003082892,0.00008073535,0.00007029482,0.00005012953,0.00006581839,0.001136457,0.0006634387,0.000038278,0.0001696713],"category_scores_gemma":[0.00002859314,0.00006096337,0.00001533377,0.0002845807,0.00003930424,0.001722893,0.0005555901,0.00006182037,0.0001596087],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005093544,"about_ca_system_score_gemma":0.00006369226,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002176749,"about_ca_topic_score_gemma":0.000006021472,"domain_scores_codex":[0.9991311,0.00002811165,0.000140897,0.000356832,0.0002214346,0.0001216476],"domain_scores_gemma":[0.9995252,0.00005092958,0.00002722217,0.0003363985,0.00003943784,0.00002079086],"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":[0.000001000464,0.0000131775,0.00001960035,0.000004873749,0.000007809864,9.845701e-7,0.0002781885,0.000006688353,0.0009505695,0.7346714,0.00974935,0.2542963],"study_design_scores_gemma":[0.0008574668,0.0002084083,0.002377952,0.0000690888,0.00002476077,0.00001584919,0.001457307,0.3705481,0.05267909,0.3454259,0.2258703,0.0004657472],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0153872,0.001591325,0.8904575,0.02687718,0.0005417303,0.0008544943,0.00000212246,0.001527107,0.06276137],"genre_scores_gemma":[0.9749982,0.00009961072,0.02235766,0.0002845736,0.00001358016,0.0001016418,0.000003482518,0.000003598169,0.002137687],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.959611,"threshold_uncertainty_score":0.9999005,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0861146672899413,"score_gpt":0.3814566730101803,"score_spread":0.2953420057202391,"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."}}