{"id":"W7106855812","doi":"10.48448/yda9-nk60","title":"Out of Sight, Not Out of Context? Egocentric Spatial Reasoning in VLMs Across Disjoint Frames","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Object (grammar); Intersection (aeronautics); Disjoint sets; Embodied cognition; Benchmark (surveying); Salient; Oracle; Representation (politics); Bottleneck","routes":{"ca_aff":true,"ca_fund":false,"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":[0.001562728,0.0008619624,0.0008031007,0.0006333116,0.0006512695,0.002238706,0.002634281,0.001723616,0.006432538],"category_scores_gemma":[0.01027982,0.0004589659,0.001104196,0.0005922503,0.001953439,0.005341122,0.003593811,0.001778213,0.001235583],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001746392,"about_ca_system_score_gemma":0.001687016,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01746769,"about_ca_topic_score_gemma":0.01580515,"domain_scores_codex":[0.9986725,0.0004063674,0.00006982529,0.0005132325,0.0002139117,0.0001241278],"domain_scores_gemma":[0.9968917,0.001624968,0.0002336369,0.0007894274,0.0002877592,0.000172646],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001576643,0.000340101,0.00889774,0.0009406781,0.0002523385,0.0007375779,0.0030214,0.3128461,0.02820136,0.1043645,0.01324765,0.5255739],"study_design_scores_gemma":[0.00008828043,0.0002421008,0.001943396,0.0001368918,0.00006446815,0.0002822459,0.0009401872,0.8216246,0.01747145,0.1446416,0.0125034,0.00006135166],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.210391,0.001388074,0.7541703,0.001496175,0.0001897458,0.0001794674,0.00189107,0.0122066,0.01808768],"genre_scores_gemma":[0.7856503,0.0002557402,0.2083391,0.0003430672,0.00002328425,0.00007463127,0.001517839,0.0004521509,0.003343916],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01746769,"threshold_uncertainty_score":0.03473198,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03700224475443828,"score_gpt":0.3458229988411389,"score_spread":0.3088207540867006,"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."}}