{"id":"W4414197327","doi":"10.1109/cvprw67362.2025.00056","title":"Revisiting Referring Expression Comprehension Evaluation in the Era of Large Multimodal Models","year":2025,"lang":"en","type":"article","venue":"","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Benchmark (surveying); Comprehension; Vocabulary; Object (grammar); Benchmarking; Key (lock)","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":[0.01825164,0.00415747,0.002249318,0.003140493,0.001622315,0.005720255,0.006097648,0.004008069,0.009941429],"category_scores_gemma":[0.07754095,0.0009652366,0.002041217,0.002917683,0.001949386,0.01037668,0.006016554,0.005548706,0.007693517],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004371594,"about_ca_system_score_gemma":0.004035451,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02300437,"about_ca_topic_score_gemma":0.02856475,"domain_scores_codex":[0.9826559,0.009724772,0.0009063719,0.003482053,0.00255797,0.0006729969],"domain_scores_gemma":[0.964476,0.02090583,0.0009038763,0.007112786,0.005719299,0.0008821291],"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":[0.001767628,0.0007283968,0.01822023,0.004458192,0.001014567,0.001183069,0.002297749,0.09700255,0.01693269,0.01189158,0.2846072,0.5598961],"study_design_scores_gemma":[0.0003340117,0.0008676931,0.01011889,0.001060098,0.0004152145,0.001075962,0.003383128,0.8101196,0.03188229,0.02376902,0.1166718,0.000302233],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.3768091,0.02019758,0.3284711,0.01037863,0.00271348,0.001596221,0.03190678,0.168227,0.05970007],"genre_scores_gemma":[0.6855817,0.002006898,0.2040181,0.00503849,0.0004729394,0.001151265,0.07717639,0.01047224,0.01408204],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02300437,"threshold_uncertainty_score":0.09652507,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04613760366182853,"score_gpt":0.3210189740561321,"score_spread":0.2748813703943035,"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."}}