{"id":"W4413144748","doi":"10.1109/cvpr52734.2025.01367","title":"Can Large Vision-Language Models Correct Semantic Grounding Errors By Themselves?","year":2025,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute","funders":"","keywords":"Computer science; Natural language processing; Ground; Artificial intelligence; Engineering","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":[0.004446526,0.002713626,0.001613342,0.0007341887,0.0008305918,0.003099296,0.004551604,0.002802928,0.004263238],"category_scores_gemma":[0.03123152,0.001078231,0.001256902,0.0005946105,0.001765629,0.007541247,0.003465303,0.005461983,0.00312124],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001374057,"about_ca_system_score_gemma":0.00233352,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008863611,"about_ca_topic_score_gemma":0.01273412,"domain_scores_codex":[0.9976153,0.0008382555,0.00009638449,0.0008573971,0.0003149209,0.0002777543],"domain_scores_gemma":[0.9911836,0.004416289,0.000511484,0.002593362,0.0009901744,0.0003050051],"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":[0.0008562849,0.0004465976,0.01099799,0.0005044751,0.0004788194,0.0004673448,0.001011984,0.3627898,0.02998983,0.01114346,0.01704271,0.5642706],"study_design_scores_gemma":[0.00006683688,0.0001577638,0.0008316053,0.00006353956,0.00006779301,0.0001495934,0.0002566986,0.9517646,0.01587198,0.02711759,0.00360548,0.00004653754],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2001363,0.001652966,0.7444264,0.003404577,0.0007172232,0.0002391253,0.0006782951,0.03990692,0.008838199],"genre_scores_gemma":[0.8279632,0.0002283423,0.1642944,0.001134501,0.00007304281,0.0001125114,0.001021771,0.001809714,0.003362478],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008863611,"threshold_uncertainty_score":0.02351576,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007488223827787919,"score_gpt":0.2949452169845906,"score_spread":0.2874569931568027,"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."}}