{"id":"W3167967639","doi":"","title":"LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs","year":2021,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Topic Modeling","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":false,"ca_institutions":"Google (Canada); University of Alberta","funders":"","keywords":"Computer science; Question answering; Knowledge graph; Hop (telecommunications); Natural language processing; Artificial intelligence; Computer network","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":[0.0004358942,0.0002013412,0.0001765941,0.000204438,0.0002332474,0.0003232043,0.0005697081,0.00007846749,0.00006520357],"category_scores_gemma":[0.000518092,0.0002088582,0.0001158664,0.0001651156,0.00001953499,0.0003051218,0.0001842987,0.0004450798,0.00005198321],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001418806,"about_ca_system_score_gemma":0.0001201859,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008068407,"about_ca_topic_score_gemma":0.00005309269,"domain_scores_codex":[0.9983137,0.0001469962,0.000323873,0.0006200853,0.0003331501,0.0002622451],"domain_scores_gemma":[0.9987898,0.0001409301,0.0001538604,0.0002908448,0.000539382,0.00008512315],"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":[0.00001942475,0.00009742856,0.003268498,0.00001397872,0.00003588141,0.0000184308,0.0005142319,0.02468551,0.003219931,0.9439581,0.00003616558,0.02413241],"study_design_scores_gemma":[0.000674614,0.00008585377,0.001736591,0.0002980365,0.000005027635,0.00001712074,0.00003491599,0.9890198,0.001792397,0.004242689,0.001872343,0.0002206022],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03293076,0.0001334831,0.9568662,0.001942843,0.001312003,0.0001574516,0.000005708362,0.0002741847,0.006377372],"genre_scores_gemma":[0.9301224,0.00005685339,0.06696106,0.0002290815,0.0001305442,0.00004116145,0.00005978507,0.00001777329,0.002381315],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9643343,"threshold_uncertainty_score":0.8516991,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08323708750218646,"score_gpt":0.3583986716908126,"score_spread":0.2751615841886262,"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."}}