{"id":"W4406075203","doi":"10.36227/techrxiv.173610007.79078639/v1","title":"HAND: Hierarchical Attention Network for Multi-Scale Handwritten Document Recognition and Layout Analysis","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Scale (ratio); Pattern recognition (psychology); Artificial intelligence; Natural language processing; Cartography; Geography","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001116485,0.0004711889,0.0008271089,0.0007767641,0.0003635438,0.001101004,0.0008190711,0.0005738173,0.00005956225],"category_scores_gemma":[0.00006317561,0.0004556445,0.0006260407,0.000781304,0.0001272142,0.0003374458,0.001658433,0.000577848,0.00001945842],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001175655,"about_ca_system_score_gemma":0.0001142538,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002574087,"about_ca_topic_score_gemma":0.0006224787,"domain_scores_codex":[0.9965619,0.0002649192,0.000787932,0.001481519,0.0003602736,0.0005434098],"domain_scores_gemma":[0.9978891,0.0002592158,0.0003161577,0.0008718218,0.0004585555,0.000205156],"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":[0.0001120438,0.0005037822,0.004206801,0.0009061934,0.00298215,0.00001157637,0.0008430851,0.0003377514,0.0002507544,0.003170158,0.009783876,0.9768918],"study_design_scores_gemma":[0.002580627,0.0003257308,0.009222973,0.001000676,0.002851005,0.00001373967,0.000055562,0.6917928,0.004333319,0.2821286,0.003685799,0.002009155],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004237412,0.0002632237,0.9905731,0.00112271,0.0004373658,0.001652036,0.0002482499,0.0007385837,0.0007272982],"genre_scores_gemma":[0.01792405,0.0002959186,0.9744456,0.0008254509,0.0002615601,0.001502586,0.001439101,0.00002168307,0.003284044],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9748827,"threshold_uncertainty_score":0.9999359,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04028039989697042,"score_gpt":0.3070075176357279,"score_spread":0.2667271177387575,"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."}}