{"id":"W7160034663","doi":"10.1109/iccv51701.2025.02155","title":"A Token-Level Text Image Foundation Model for Document Understanding","year":2025,"lang":"","type":"article","venue":"","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Natural Science Foundation of China","keywords":"Foundation (evidence); Image (mathematics); Context (archaeology); Image processing; Feature (linguistics)","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","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001095808,0.0004122523,0.0004044894,0.0006338329,0.00068381,0.001780948,0.0009901007,0.000222658,0.0003731054],"category_scores_gemma":[0.0001510467,0.0004352823,0.0002779361,0.0008448522,0.0001619506,0.002204453,0.0005693717,0.0002249087,0.0001265437],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001299879,"about_ca_system_score_gemma":0.0007255895,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004423764,"about_ca_topic_score_gemma":0.00006869308,"domain_scores_codex":[0.9968786,0.00009350346,0.0008666337,0.001019637,0.0004288287,0.0007127573],"domain_scores_gemma":[0.9980389,0.0003268713,0.0002306263,0.0007715512,0.0004793289,0.0001527457],"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":[0.00003721936,0.0001387893,0.000004505726,0.000198936,0.00007720587,0.000001552545,0.0006075819,0.00007158111,0.001362586,0.7310404,0.0162611,0.2501985],"study_design_scores_gemma":[0.0007181778,0.00008809599,0.000005332144,0.0001905157,0.0000414071,0.000002244817,0.0001542818,0.5819718,0.01312215,0.4025213,0.000888991,0.0002957733],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0000487921,0.00007123517,0.9436703,0.007080187,0.0005913447,0.001905764,0.00002360776,0.000532264,0.04607647],"genre_scores_gemma":[0.3996665,0.000149432,0.5700198,0.002009861,0.0000539954,0.0003536002,0.00001998258,0.0000227326,0.02770419],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5819002,"threshold_uncertainty_score":0.9998099,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09104862361110418,"score_gpt":0.3344530283190299,"score_spread":0.2434044047079257,"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."}}