{"id":"W4410491803","doi":"10.1109/jstsp.2025.3558652","title":"R-Bench: Are Your Large Multimodal Model Robust to Real-World Corruptions?","year":2025,"lang":"en","type":"article","venue":"IEEE Journal of Selected Topics in Signal Processing","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"National Natural Science Foundation of China","keywords":"Computer science; Artificial intelligence; Computer vision","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005683632,0.0001729954,0.0003172632,0.0009648846,0.0001685271,0.0002572036,0.001098236,0.0001042118,0.000005084862],"category_scores_gemma":[0.0001257527,0.0001681274,0.00005303051,0.002535368,0.00002719521,0.0009928339,0.0001071308,0.0006132442,0.000002408148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003343363,"about_ca_system_score_gemma":0.0006902778,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000171447,"about_ca_topic_score_gemma":0.00009468114,"domain_scores_codex":[0.9981051,0.00007962409,0.0007639116,0.000300222,0.0003945697,0.0003565681],"domain_scores_gemma":[0.9979955,0.00005021896,0.0005060314,0.0002823074,0.001063349,0.0001026102],"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":[0.0003447011,0.002579571,0.0585973,0.0007386028,0.0001510542,0.0002991877,0.005485286,0.338311,0.1455571,0.04725592,0.04240631,0.3582739],"study_design_scores_gemma":[0.0006391781,0.0000734666,0.01290618,0.0007440806,0.00001511214,0.00002367753,0.00006930951,0.9606959,0.01555332,0.007826787,0.001173612,0.0002793786],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01355096,0.00009868684,0.9823399,0.00240062,0.0001858427,0.0001500136,0.000005630989,0.0001079076,0.001160394],"genre_scores_gemma":[0.7389536,0.00003551708,0.2595075,0.0005772716,0.0001300463,0.000009292622,0.000001781179,0.000009664502,0.000775239],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7254027,"threshold_uncertainty_score":0.6856037,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04899085643834207,"score_gpt":0.3323187745233619,"score_spread":0.2833279180850199,"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."}}