{"id":"W7131095609","doi":"10.1109/iccvw69036.2025.00782","title":"DIVE-Doc: Downscaling Foundational Image Visual Encoder into Hierarchical Architecture for DocVQA","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":"École de Technologie Supérieure","funders":"","keywords":"Encoder; Distillation; Image (mathematics); Code (set theory); Architecture; Forcing (mathematics); Visualization","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","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0009295727,0.0006038227,0.0005475817,0.0005885296,0.001632988,0.001287853,0.002096487,0.0003080841,0.0005037093],"category_scores_gemma":[0.0008658842,0.0005738598,0.0004404119,0.001364051,0.0007883506,0.0005061824,0.00150841,0.001229513,0.0002599783],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002378159,"about_ca_system_score_gemma":0.000864368,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00114238,"about_ca_topic_score_gemma":0.0001534322,"domain_scores_codex":[0.99522,0.0003311054,0.001015036,0.001795013,0.0006996749,0.0009391438],"domain_scores_gemma":[0.9955763,0.00218854,0.0002477955,0.001103828,0.0005406053,0.0003429678],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001184409,0.0005842585,0.001632923,0.0002586099,0.0001853072,0.000002971476,0.001926591,0.008915856,0.006913627,0.583182,0.001643502,0.3946359],"study_design_scores_gemma":[0.001307286,0.000155105,0.01006892,0.0001017856,0.00006627353,0.000009479687,0.00004803015,0.8273973,0.001042334,0.1339159,0.02530994,0.0005775594],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01030363,0.0001134037,0.9176798,0.06234746,0.0008716757,0.001521807,0.00002458492,0.000363771,0.006773873],"genre_scores_gemma":[0.5230602,0.000006886087,0.4717912,0.002091038,0.0004057877,0.0003919454,0.00006752516,0.00002852803,0.002156908],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8184815,"threshold_uncertainty_score":0.9997489,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008780846628456837,"score_gpt":0.339472735249167,"score_spread":0.3306918886207101,"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."}}