{"id":"W7160112519","doi":"10.1109/iccv51701.2025.00687","title":"Physics Context Builders: A Modular Framework for Physical Reasoning in Vision-Language Models","year":2025,"lang":"","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Canadian Institute for Advanced Research","keywords":"Context (archaeology); Modular design; Automated reasoning; Physical system; Knowledge representation and reasoning","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005187527,0.0005044421,0.000697959,0.0002314386,0.0003962662,0.0005452982,0.001679989,0.0002833195,0.0000164636],"category_scores_gemma":[0.0005606337,0.0005286229,0.0003414759,0.001842039,0.0001649267,0.0007939067,0.0007717625,0.001076519,0.00005041714],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002037836,"about_ca_system_score_gemma":0.0003111078,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001344467,"about_ca_topic_score_gemma":0.00004372697,"domain_scores_codex":[0.9963958,0.0002153658,0.0006182816,0.00151213,0.0004181631,0.0008402748],"domain_scores_gemma":[0.9959964,0.001586776,0.0002301918,0.001790087,0.0002218809,0.0001747082],"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":[0.00002622772,0.0004778944,0.00007504213,0.00006217775,0.00003332601,0.000001855554,0.005503271,0.2147885,0.0001661024,0.5079964,0.0000871295,0.2707821],"study_design_scores_gemma":[0.000798014,0.0001111006,0.0003962975,0.000627922,0.00002677101,6.766971e-7,0.0004366922,0.7166174,0.0003721551,0.2801959,0.00006907406,0.0003479807],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02949471,0.000535662,0.96179,0.00437222,0.000275493,0.001611757,0.00001791032,0.0002504133,0.001651839],"genre_scores_gemma":[0.6051801,0.000009109139,0.3931012,0.0009129953,0.0001656254,0.000298611,0.000006938093,0.00002700646,0.0002984487],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5756854,"threshold_uncertainty_score":0.9997165,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01607597050264624,"score_gpt":0.3503052142448334,"score_spread":0.3342292437421871,"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."}}