{"id":"W3132507547","doi":"10.1016/j.patcog.2022.109209","title":"Visual question answering from another perspective: CLEVR mental rotation tests","year":2022,"lang":"en","type":"article","venue":"Pattern Recognition","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"Canadian Institute for Advanced Research; McGill University; Minnow Environmental (Canada); Polytechnique Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Institut de Valorisation des Données; Natural Sciences and Engineering Research Council of Canada; Mitacs; Artificial Intelligence Research Center; Canadian Institute for Advanced Research","keywords":"Mental rotation; Perspective (graphical); Artificial intelligence; Computer science; Mental image; Rotation (mathematics); Perception; Object (grammar); Question answering; Computer vision; Visual perception; Cognitive psychology; Psychology; Cognition","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003466555,0.001390627,0.0007454132,0.001787642,0.0005558708,0.002779163,0.00187841,0.002806428,0.0194506],"category_scores_gemma":[0.04930216,0.0002382676,0.0007466698,0.0007452907,0.001442826,0.005220779,0.001690284,0.001748695,0.003091828],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007377054,"about_ca_system_score_gemma":0.0006303227,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003724637,"about_ca_topic_score_gemma":0.002249807,"domain_scores_codex":[0.996348,0.001707943,0.0003253824,0.0007018848,0.0007050319,0.0002117847],"domain_scores_gemma":[0.9674434,0.0242565,0.001557909,0.00428278,0.001635951,0.0008234686],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00964108,0.005250419,0.1003044,0.001870218,0.000877372,0.001986824,0.007697322,0.01520255,0.05412023,0.04644903,0.09706356,0.659537],"study_design_scores_gemma":[0.003844994,0.009662379,0.3887379,0.0009072694,0.000900807,0.009034367,0.009938846,0.1549845,0.09826171,0.1658631,0.1568981,0.0009659471],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8573987,0.002648584,0.03433108,0.003128498,0.0006499589,0.001000979,0.005080454,0.001927956,0.09383382],"genre_scores_gemma":[0.9741516,0.0003279844,0.01024563,0.0008414846,0.0001693116,0.0004280806,0.004084342,0.0001851321,0.009566457],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0194506,"threshold_uncertainty_score":0.06506878,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02354854623864443,"score_gpt":0.3167079208667227,"score_spread":0.2931593746280782,"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."}}