{"id":"W4389520485","doi":"10.18653/v1/2023.findings-emnlp.652","title":"NEWTON: Are Large Language Models Capable of Physical Reasoning?","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Office of Naval Research; Natural Sciences and Engineering Research Council of Canada; Defense Advanced Research Projects Agency; National Science Foundation","keywords":"Benchmark (surveying); Computer science; Qualitative reasoning; Consistency (knowledge bases); Automated reasoning; Artificial intelligence; Mainstream; Natural language processing; Data science","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":[],"consensus_categories":[],"category_scores_codex":[0.000149031,0.00006847851,0.0001329038,0.00005745148,0.00003627874,0.00002721159,0.0004554342,0.00002748861,0.00001329317],"category_scores_gemma":[0.00001991461,0.00005875725,0.00005270634,0.0003489627,0.000009192384,0.0002720557,0.0002758732,0.00006520745,0.00007845054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001075764,"about_ca_system_score_gemma":0.00002442181,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001341379,"about_ca_topic_score_gemma":0.00001854204,"domain_scores_codex":[0.9992033,0.00001904839,0.0001037799,0.000229678,0.0002083744,0.0002357666],"domain_scores_gemma":[0.9993411,0.00003799214,0.00004208768,0.0004962823,0.00003322642,0.00004931545],"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":[0.000001631384,0.00007395535,0.000425625,0.00003529634,0.00001369534,0.0000392564,0.007331512,0.0299903,0.001837271,0.9503257,0.005280285,0.004645534],"study_design_scores_gemma":[0.0001243305,0.000009990044,0.0001713102,0.0000167571,0.000001510379,0.000001281182,0.0002830212,0.9902349,0.002873384,0.005955761,0.0002577542,0.0000700157],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2795272,0.00003714068,0.7054073,0.0005562209,0.00009209133,0.00006633109,0.000003153113,0.0004323725,0.01387815],"genre_scores_gemma":[0.9829202,0.000001731613,0.01224567,0.0001178875,0.00006779475,0.000004616752,0.000001421169,0.000006258359,0.004634456],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9602446,"threshold_uncertainty_score":0.2396051,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02060342473632154,"score_gpt":0.2654812464257332,"score_spread":0.2448778216894117,"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."}}