{"id":"W4405312339","doi":"10.48550/arxiv.2412.08619","title":"Physics Context Builders: A Modular Framework for Physical Reasoning in Vision-Language Models","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Robotics and Automated Systems","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Government of Canada; Canadian Institute for Advanced Research","keywords":"Training (meteorology); Computer science; Artificial intelligence; Physics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001539445,0.002027121,0.0009336206,0.001217276,0.0007567061,0.002763691,0.004935114,0.001613498,0.007259087],"category_scores_gemma":[0.005241882,0.001578189,0.003796336,0.0006183544,0.001544173,0.004587802,0.00486388,0.005080172,0.00327946],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001504037,"about_ca_system_score_gemma":0.002346439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009853521,"about_ca_topic_score_gemma":0.02895133,"domain_scores_codex":[0.9991499,0.0001937992,0.00006964866,0.0002681667,0.000244323,0.00007413177],"domain_scores_gemma":[0.9987106,0.0005226763,0.00008186575,0.0004434459,0.0001458293,0.0000955107],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003305654,0.0004195623,0.00287412,0.001002087,0.0004026032,0.0004920388,0.000925488,0.4334973,0.01675216,0.1569277,0.03822955,0.3481468],"study_design_scores_gemma":[0.00005199517,0.00004365599,0.0001746193,0.00005987086,0.00004694717,0.00007760761,0.00005349681,0.8821799,0.003725694,0.09925846,0.01429396,0.00003380072],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00317612,0.0002667054,0.9701335,0.0004163828,0.00006685381,0.0001427744,0.0007816781,0.0232516,0.001764383],"genre_scores_gemma":[0.1466999,0.0005110073,0.8448191,0.0005585966,0.00009275212,0.0004722483,0.002679549,0.001789953,0.002376952],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009853521,"threshold_uncertainty_score":0.02428412,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0444426252677881,"score_gpt":0.2118638709412499,"score_spread":0.1674212456734618,"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."}}