{"id":"W2781014390","doi":"10.1145/3086512.3086550","title":"Two-step cascaded textual entailment for legal bar exam question answering","year":2017,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"National Institute of Informatics; Alberta Machine Intelligence Institute","keywords":"Textual entailment; Computer science; Logical consequence; Natural language processing; Artificial intelligence; Question answering; Negation; Information retrieval; Representation (politics); Programming language","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003220153,0.00009924708,0.0001039097,0.00003355022,0.0003354991,0.000473383,0.0008442938,0.00003814272,0.00001582156],"category_scores_gemma":[0.00004847561,0.00009015703,0.00004933092,0.00001593704,0.00002007762,0.0008301941,0.0002913777,0.00006655209,0.00002657116],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005668892,"about_ca_system_score_gemma":0.00003744738,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007761434,"about_ca_topic_score_gemma":0.0001502063,"domain_scores_codex":[0.9990582,0.00001841376,0.0001657638,0.0003414627,0.0001720801,0.0002440847],"domain_scores_gemma":[0.9989995,0.00003026848,0.00007424532,0.0007766307,0.00004416862,0.00007521146],"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":[0.000007015927,0.00003321899,0.000704564,0.00001663792,0.00001389714,0.00001048869,0.0002180942,0.0008028403,0.003067746,0.8575476,0.0007052326,0.1368726],"study_design_scores_gemma":[0.0009175664,0.00008350962,0.001723777,0.00003019094,0.000006751693,0.00002006521,0.00007005792,0.9706181,0.01031479,0.00100244,0.01496787,0.0002448507],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03329245,0.00001721468,0.9547529,0.001199565,0.0005308245,0.0002217708,7.875692e-7,0.0001358409,0.009848658],"genre_scores_gemma":[0.7677224,0.000001593301,0.2301562,0.0001575855,0.000172033,0.00002367115,8.033427e-7,0.000005260008,0.001760447],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9698153,"threshold_uncertainty_score":0.4564843,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03522215394083796,"score_gpt":0.3092648581678765,"score_spread":0.2740427042270385,"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."}}