{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002911944,0.0001583341,0.0002275097,0.0009879516,0.000225946,0.00006548627,0.0004403074,0.0002303748,0.00000225113],"category_scores_gemma":[0.0000787974,0.0001561798,0.00005217237,0.0009215378,0.0000788598,0.00008929672,0.000008333551,0.0004277152,0.000005539091],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008997117,"about_ca_system_score_gemma":0.0000601524,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006682552,"about_ca_topic_score_gemma":0.0001477211,"domain_scores_codex":[0.9987634,0.00005590972,0.0003560939,0.0004593587,0.00009479637,0.000270457],"domain_scores_gemma":[0.9990926,0.0003045253,0.00005151628,0.000454366,0.00005507208,0.00004187923],"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":[0.00005327688,0.00040947,0.003419664,0.0003275947,0.00007031605,0.000005270366,0.0002742547,0.03173786,0.005593534,0.3417804,0.000009899821,0.6163184],"study_design_scores_gemma":[0.000613989,0.0002648819,0.001409747,0.0005074831,0.00002363867,0.00002630737,0.0001529713,0.9702283,0.01804575,0.005884711,0.002590789,0.000251397],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0668355,0.0001863184,0.9255808,0.006146681,0.0001950646,0.0006617692,0.000003401836,0.0003329319,0.00005753616],"genre_scores_gemma":[0.9790961,0.00003489834,0.02001721,0.00005831306,0.000008409958,0.0006682436,0.000002602805,0.00001119411,0.0001030147],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9384905,"threshold_uncertainty_score":0.6368825,"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."}}