{"id":"W4409539127","doi":"10.1145/3729242","title":"Cascade Transformer for Hierarchical Semantic Reasoning in Text-Based Visual Question Answering","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Intelligent Systems and Technology","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"St. Francis Xavier University","funders":"National Natural Science Foundation of China","keywords":"Computer science; Question answering; Transformer; Cascade; Natural language processing; Artificial intelligence; Information retrieval","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":[0.001045076,0.001206602,0.0007105925,0.001232919,0.0004222804,0.001006461,0.002438155,0.001320497,0.006626219],"category_scores_gemma":[0.002895039,0.0004456959,0.001862869,0.0007225037,0.0007091211,0.00354756,0.00144949,0.001709654,0.002015973],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001466869,"about_ca_system_score_gemma":0.001095899,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01034839,"about_ca_topic_score_gemma":0.01319801,"domain_scores_codex":[0.9993399,0.0001087391,0.00004226893,0.0002893128,0.0001402707,0.00007959183],"domain_scores_gemma":[0.99944,0.0002552972,0.00004231119,0.00008778939,0.0001334197,0.00004118806],"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":[0.0007364787,0.0005204321,0.001849708,0.0007295528,0.000185078,0.0006569778,0.001006426,0.08727907,0.05912096,0.02957004,0.02054867,0.7977966],"study_design_scores_gemma":[0.00004082144,0.0001172895,0.0006012848,0.0000394933,0.0001027247,0.0001968365,0.0001297736,0.9248965,0.01818934,0.04944407,0.006206356,0.00003547544],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02068416,0.00089852,0.9647592,0.0003860768,0.0001070212,0.0003072736,0.0008050872,0.008257406,0.003795258],"genre_scores_gemma":[0.5885856,0.0008149458,0.3974293,0.0008233868,0.0001176702,0.0003886984,0.003694637,0.0003121585,0.007833611],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01034839,"threshold_uncertainty_score":0.02216691,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01049179315655855,"score_gpt":0.308018523082175,"score_spread":0.2975267299256164,"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."}}