{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001180351,0.0003870322,0.0005357662,0.0001664656,0.0000473443,0.0001160169,0.0003995941,0.0003984794,0.000003539961],"category_scores_gemma":[0.00001894788,0.0004528397,0.0003093687,0.0003692883,0.00004078321,0.0001240634,0.00036091,0.0009270497,0.00003661574],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002509914,"about_ca_system_score_gemma":0.00004836429,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007423539,"about_ca_topic_score_gemma":0.00001593425,"domain_scores_codex":[0.9986023,0.00003661172,0.0002115695,0.0006766251,0.00008027224,0.0003926196],"domain_scores_gemma":[0.9990541,0.0001288814,0.00006787855,0.0005958111,0.00005058758,0.0001027741],"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":[0.00001230219,0.00004469523,0.000007102822,0.0006087348,0.0001353879,0.0001518977,0.001466457,0.8881232,0.00009034597,0.1083996,0.0001238773,0.0008364546],"study_design_scores_gemma":[0.0002333947,0.00002236184,0.000003730298,0.001420198,0.00007256831,8.791004e-7,0.0005324701,0.7963576,0.0001166818,0.2008541,0.00001634254,0.0003697006],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2916394,0.0005331095,0.7056837,0.00001080049,0.0006031286,0.0004526976,0.00007838199,0.0004830442,0.0005157096],"genre_scores_gemma":[0.9960599,0.00007360621,0.003238601,0.00001351827,0.0003018925,0.000006884297,0.00004124265,0.0001101981,0.0001541694],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7044204,"threshold_uncertainty_score":0.9997923,"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."}}