{"id":"W2968398601","doi":"10.18653/v1/d19-1458","title":"CLUTRR: A Diagnostic Benchmark for Inductive Reasoning from Text","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute; McGill University","funders":"Samsung; Samsung Advanced Institute of Technology; Canadian Institute for Advanced Research","keywords":"Robustness (evolution); Computer science; Natural language understanding; Inductive logic programming; Artificial intelligence; Suite; Benchmark (surveying); Generalization; Machine learning; Theoretical computer science; Natural language; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007734243,0.002531002,0.001073534,0.007350564,0.001505546,0.00342971,0.007348122,0.003881194,0.01394591],"category_scores_gemma":[0.06554406,0.00092105,0.001846925,0.004802222,0.001697244,0.006072996,0.004626847,0.002966356,0.008088778],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002368789,"about_ca_system_score_gemma":0.003588469,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008758047,"about_ca_topic_score_gemma":0.01076524,"domain_scores_codex":[0.9885283,0.004439854,0.001435738,0.0018364,0.003221148,0.0005384171],"domain_scores_gemma":[0.9425637,0.043989,0.001350498,0.00588793,0.005149227,0.001059727],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002454782,0.001829507,0.007801753,0.006759235,0.0005282257,0.001252225,0.001104186,0.06099462,0.007050586,0.03903824,0.3574414,0.5137452],"study_design_scores_gemma":[0.001643577,0.0005708609,0.003075215,0.0007555136,0.0002423046,0.001134695,0.001183265,0.6913245,0.03356064,0.1191423,0.1472154,0.0001517358],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1080913,0.01044301,0.4680173,0.006801855,0.001779381,0.00256126,0.1035378,0.2504634,0.04830462],"genre_scores_gemma":[0.2419358,0.001296958,0.6049912,0.000997317,0.0002612272,0.001375565,0.1354756,0.006686255,0.006980083],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01394591,"threshold_uncertainty_score":0.04665369,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02735789800130105,"score_gpt":0.2664366112061615,"score_spread":0.2390787132048605,"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."}}