{"id":"W4416252247","doi":"10.1109/ijcnn64981.2025.11228324","title":"GNN-ViTCap: GNN-Enhanced Multiple Instance Learning with Vision Transformers for Whole Slide Image Classification and Captioning","year":2025,"lang":"","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Pattern recognition (psychology); Cluster analysis; Closed captioning; Feature extraction; Contextual image classification; Graph; Feature vector; Pixel; Artificial neural network","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":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.0008450664,0.0005391047,0.000513653,0.0004200148,0.001797809,0.0009226211,0.0006285237,0.0002492603,0.00002479125],"category_scores_gemma":[0.000367071,0.000520997,0.0001327173,0.0014248,0.0003521652,0.001529792,0.000156383,0.0008741351,0.00004299213],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002402156,"about_ca_system_score_gemma":0.0003190614,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004312631,"about_ca_topic_score_gemma":0.0003299008,"domain_scores_codex":[0.9961097,0.0002084427,0.0008175846,0.001664741,0.0004275122,0.0007719797],"domain_scores_gemma":[0.9973314,0.0008218514,0.0004274666,0.0006321511,0.0005555413,0.0002316196],"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":[0.0004125966,0.0003338223,0.006976148,0.0005585971,0.0001333251,0.000001469065,0.003744667,0.01314582,0.2786806,0.0321586,0.0001904329,0.663664],"study_design_scores_gemma":[0.002942632,0.0003963891,0.05599693,0.0004659741,0.00007469492,0.000004542558,0.001073837,0.9243299,0.008165,0.0004298897,0.00553898,0.0005812611],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1303747,0.0002334382,0.8533223,0.006763181,0.0001625913,0.001838969,0.00001174876,0.0003650894,0.006927998],"genre_scores_gemma":[0.80923,0.0000759899,0.1871751,0.0002793434,0.00004429201,0.0004630051,0.00004435773,0.00004297452,0.002645022],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9111841,"threshold_uncertainty_score":0.9997241,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00910167173983274,"score_gpt":0.2866346593070621,"score_spread":0.2775329875672293,"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."}}