{"id":"https://openalex.org/W7225423610","doi":"https://doi.org/10.5281/zenodo.24023071","title":"Line reading order from region types for modern Hebrew document pages, trained from scratch on synthetic pages; reading-order model of the HeOCR R4d + A1 + R3h pipeline","display_name":"Line reading order from region types for modern Hebrew document pages, trained from scratch on synthetic pages; reading-order model of the HeOCR R4d + A1 + R3h pipeline","publication_year":2026,"publication_date":"2026-11-07","ids":{"openalex":"https://openalex.org/W7225423610","doi":"https://doi.org/10.5281/zenodo.24023071"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24023071","doi":"https://doi.org/10.5281/zenodo.24023071","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24023071","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24023071","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161045255","display_name":"Shay Palachy Affek","orcid":"https://orcid.org/0000-0002-1459-9320","observed_orcids":["https://orcid.org/0000-0002-1459-9320"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Palachy Affek, Shay","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-1459-9320","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5161072239","display_name":"HeOCR maintainers","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"HeOCR maintainers","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T14339","display_name":"Image Processing and 3D Reconstruction","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12707","display_name":"Vehicle License Plate Recognition","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/reading-order","display_name":"reading order","score":0.7829999923706055},{"id":"https://openalex.org/keywords/synthetic-training-data","display_name":"synthetic training data","score":0.6620000004768372},{"id":"https://openalex.org/keywords/kraken","display_name":"Kraken","score":0.4180000126361847},{"id":"https://openalex.org/keywords/document-image-analysis","display_name":"document image analysis","score":0.5230000019073486},{"id":"https://openalex.org/keywords/optical-character-recognition","display_name":"optical character recognition","score":0.5709999799728394},{"id":"https://openalex.org/keywords/document-segmentation","display_name":"document segmentation","score":0.335999995470047},{"id":"https://openalex.org/keywords/digitization-of-historical-documents","display_name":"digitization of historical documents","score":0.42899999022483826},{"id":"https://openalex.org/keywords/training-from-scratch","display_name":"training from scratch","score":0.3409999907016754}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6915000081062317},{"id":"https://openalex.org/C2781235140","wikidata":"https://www.wikidata.org/wiki/Q275131","display_name":"Scratch","level":2,"score":0.6437000036239624},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6417999863624573},{"id":"https://openalex.org/C554936623","wikidata":"https://www.wikidata.org/wiki/Q199657","display_name":"Reading (process)","level":2,"score":0.6133999824523926},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.5239999890327454},{"id":"https://openalex.org/C198352243","wikidata":"https://www.wikidata.org/wiki/Q37105","display_name":"Line (geometry)","level":2,"score":0.5023000240325928},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.4823000133037567},{"id":"https://openalex.org/C91304198","wikidata":"https://www.wikidata.org/wiki/Q9288","display_name":"Hebrew","level":2,"score":0.43299999833106995},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.42570000886917114},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4090999960899353},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.39489999413490295},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C154504017","wikidata":"https://www.wikidata.org/wiki/Q853614","display_name":"Identifier","level":2,"score":0.34060001373291016},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3057999908924103},{"id":"https://openalex.org/C2385561","wikidata":"https://www.wikidata.org/wiki/Q45432","display_name":"RSS","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2851000130176544},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.2759999930858612},{"id":"https://openalex.org/C183469790","wikidata":"https://www.wikidata.org/wiki/Q333501","display_name":"Crash","level":2,"score":0.25360000133514404},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.25279998779296875},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.25099998712539673}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24023071","doi":"https://doi.org/10.5281/zenodo.24023071","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24023071","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},{"id":"doi:10.5281/zenodo.24023072","doi":"https://doi.org/10.5281/zenodo.24023072","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24023072","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.5281/zenodo.24023071","doi":"https://doi.org/10.5281/zenodo.24023071","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24023071","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7225377301","https://openalex.org/W7225417216","https://openalex.org/W7163690831","https://openalex.org/W7167052503","https://openalex.org/W7167499215","https://openalex.org/W3208193076","https://openalex.org/W7172106642","https://openalex.org/W4210271679","https://openalex.org/W7223032124","https://openalex.org/W7047328609"],"abstract_inverted_index":{"HeOCR":[0,43,147,1202],"A1:":[1,567],"Hebrew":[2,22,418,437],"document":[3,23,419,636],"line":[4,171,181,218,241,273,336,354,377,404,454,556,582,618,702,768,1125,1297],"reading":[5,74,128,213,434,577],"order":[6,129,182,214,242,342,412,435,578,601,619,953],"(kraken)":[7],"A1":[8,52,70,197,204,223,425,446,479,590,766,973,1038,1058,1069],"is":[9,53,97,105,659,741,824,940,946,962,1126],"a":[10,20,98,239,249,277,301,349,352,486,616,807,825,941,1029,1153],"kraken":[11,135,200,229,257,384,483,496,518,562,595,1190,1213],"reading-order":[12,55,274,851],"model.":[13],"It":[14,96,285,812,870,1142,1186,1211],"orders":[15,1182],"the":[16,54,64,159,216,231,245,305,356,376,381,387,395,399,422,476,533,540,547,554,559,580,690,727,760,773,834,855,861,877,884,899,915,925,1044,1104,1120,1129,1160,1208,1215,1232,1238,1245,1347],"text":[17,236,531,613,639,740,774,1042,1046,1276],"lines":[18,65,368,415,543],"of":[19,57,102,288,440,488,539,798,809,896,914,968,1032,1107,1113,1139],"modern":[21,417],"page":[24,525,788],"from":[25,46,158,183,248,351,649,726,781,830,860],"each":[26,91,184,292,656,1291],"line's":[27,185,306,657],"position":[28],"and":[29,66,151,298,300,452,466,537,558,589,633,645,708,733,757,770,778,806,867,886,908,1034,1050,1098,1122,1195,1223,1244,1253,1279,1314,1318,1330],"its":[30,103,211,235,294,660,1295,1341],"region":[31,186,227,265,307,328,360,383,401,450,506,572,661,1146,1150,1292],"type:":[32,308],"title,":[33],"body,":[34],"stamp,":[35],"signature,":[36],"footer,":[37],"margin,":[38],"identifier":[39],"or":[40,345,637,921],"table":[41],"cell.":[42],"trained":[44,137,225,333,920],"it":[45,362,939,1183],"scratch":[47],"on":[48,106,112,226,416,436,448,520,522,833,841,936,1047,1144,1159,1334],"synthetic":[49,107,1168,1234],"pages":[50,648,688,965],"only.":[51],"step":[56],"HeOCR's":[58],"three-model":[59],"pipeline:":[60],"-":[61,69,76,210,234,568,611],"R4d":[62,374,423,588,1067],"finds":[63],"their":[67],"regions;":[68],"puts":[71],"them":[72],"in":[73,405,459,526,620,696,719,952,1136,1175,1308],"order;":[75],"R3h":[77,427,916,1071,1074],"reads":[78],"them.":[79],"The":[80,94,164,330,530,687,722,739,791,910,1167],"three":[81],"are":[82,571,718,912,966,1110,1134,1165,1247],"released":[83],"together,":[84],"as":[85,120,1294,1332],"separate":[86],"records":[87],"that":[88,655,873,880],"link":[89],"to":[90,220,472,515,759,1054,1323],"other":[92,443],"(see":[93,149,462,1192],"pipeline).":[95],"research":[99,430],"baseline.":[100],"All":[101,628,944],"evaluation":[104,111],"pages;":[108],"no":[109,634,754,850,1241],"reviewed":[110],"real":[113,635,1178],"scans":[114],"exists":[115],"yet.":[116],"File":[117],"heocr-a1-reading-order.safetensors":[118,263,504],"(trained":[119],"best_0.0000.safetensors),":[121],"8,492":[122],"bytes":[123],"sha256":[124],"69aa1946583bb68ddf13ef20860582f4aa982a46b2f79e38bd7d9ae20535fb56":[125],"Model":[126,167],"type":[127,187,397,1298],"(line":[130],"level,":[131],"--level":[132,789],"baselines)":[133],"Format":[134],"safetensors;":[136],"with":[138,280,334,475,517,599,753,803,957,1189,1250,1325],"ketos":[139,584,784,865,878,1300],"7.0.2":[140,201,461,519,563,785,1191,1301],"rotrain":[141,786],"Licence":[142,1194],"Apache-2.0,":[143],"Copyright":[144,1199],"(c)":[145,1200],"2026":[146,1201],"maintainers":[148],"LICENSE":[150],"NOTICE)":[152],"Training":[153,624,626,776,846,1257,1275,1299,1304],"record":[154,856,1305],"a1_training_run.json":[155,1306],"(heocr_engine_training_run.v1,":[156],"reconstructed":[157],"run's":[160,862],"evidence)":[161],"Code":[162],"HeOCR/heocr-engine":[163],"pipeline":[165,247,458],"Step":[166],"Record":[168],"Segmentation:":[169],"baselines,":[170],"polygons,":[172],"typed":[173,375],"regions":[174],"R4d,":[175],"heocr-r4d-segmentation.safetensors":[176,261,501],"(d32cd1a8\u2026)":[177],"10.5281/zenodo.24023070":[178],"Reading":[179,411],"order:":[180],"A1,":[188,221],"this":[189,1309,1338],"model":[190,477,602,1239,1339],"10.5281/zenodo.24023072":[191],"Recognition":[192],"R3h,":[193],"heocr-r3h-recognition.safetensors":[194,267,509],"(f76affa8\u2026)":[195],"10.5281/zenodo.24023074":[196],"needs":[198,1187],"heocr-engine.":[199],"cannot":[202,565,875],"run":[203,490],"by":[205,546,744,795,1340],"itself,":[206],"for":[207,291,386,402,413,432,751],"two":[208,650],"reasons:":[209],"neural":[212,576],"passes":[215],"segmenter's":[217,581],"types":[219,337,361,451],"while":[222],"was":[224,631,640,857,919,934,1039,1157],"types.":[228,329,583,1147],"rejects":[230],"class-mapping":[232],"mismatch;":[233],"output":[237,614],"ignores":[238,615],"learned":[240,433,548,617],"(mittagessen/kraken#820).":[243,623],"Run":[244],"whole":[246],"heocr-engine":[250,256,365,495,848,1188,1286,1316],"checkout":[251,487],"(pip":[252],"install":[253,492],"'.[kraken]'):":[254],"sh":[255],"--api":[258,497],"python":[259,498],"--segmentation-model":[260,500],"--reading-order-model":[262,503],"--reading-order-classes":[264,505],"--recognition-model":[266,508],"page.jpg":[268,511],"page.json":[269,512],"Architecture":[270],"Network:":[271],"kraken's":[272,456,575,612,971],"network":[275,332],"(ROMLP),":[276],"two-layer":[278],"perceptron":[279],"about":[281,527],"1.9":[282],"K":[283],"parameters.":[284],"compares":[286],"pairs":[287],"lines.":[289,959],"Inputs:":[290],"line,":[293],"baseline's":[295],"start,":[296],"end":[297,514,516,1055,1324],"centre,":[299],"one-hot":[302],"class.":[303,1155],"Classes:":[304],"default":[309],"0,":[310],"title":[311],"1,":[312],"body":[313,353,1124],"2,":[314],"stamp_area":[315,372],"3,":[316],"signature_area":[317],"4,":[318],"footer":[319],"5,":[320],"margin":[321],"6,":[322],"identifier_area":[323],"7,":[324],"table_cells":[325],"8.":[326],"Why":[327],"same":[331,357],"R4d's":[335,367,414,449,663,958,1145],"(default,":[338],"stamp)":[339],"could":[340],"not":[341,607,772,882,1248],"archive":[343,1101],"cards":[344,1033,1109,1133],"ledgers.":[346],"Nothing":[347,630],"told":[348],"label":[350],"at":[355,892],"height.":[358],"With":[359],"can.":[363],"How":[364,471],"tags":[366],"(--reading-order-classes":[369],"region):":[370],"1.":[371],"if":[373],"stamp;":[378],"2.":[379],"otherwise":[380,390],"first":[382,1130],"lists":[385,871],"line;":[388],"3.":[389],"default.":[391],"On":[392,1027,1103],"matched":[393,532,955],"lines,":[394,956],"predicted":[396,560],"equalled":[398],"gold":[400,945],"every":[403,1048],"all":[406],"four":[407],"batches":[408,723,911,1161,1258],"checked.":[409],"Uses":[410],"pages,":[420],"inside":[421],"+":[424,426,735,1057,1059,1068,1070,1073,1282],"pipeline.":[428],"A":[429,1148],"baseline":[431],"documents.":[438],"Out":[439],"scope":[441],"Segmenters":[442],"than":[444],"R4d.":[445],"depends":[447,1143],"stamp":[453],"type.":[455,662],"own":[457,1296],"version":[460,1342],"below).":[463],"Real":[464],"scans,":[465],"layouts":[467],"unlike":[468],"hocrsyngen's":[469,1171],"templates.":[470],"get":[473],"started":[474],"Use":[478],"through":[480],"heocr-engine's":[481],"in-process":[482],"stage.":[484],"In":[485],"heocr-engine,":[489],"pip":[491],"'.[kraken]',":[493],"then:":[494],"\\":[499,502,507,510],"Tested":[513],"CPU,":[521],"one":[523,592,1117],"seed-101":[524,974,1075],"8":[528,538],"s.":[529],"scored":[534],"prediction":[535],"exactly,":[536],"page's":[541],"10":[542,805,1106],"were":[544,724,793,881,1173],"moved":[545],"order.":[549,969,1114],"--geometry-output":[550],"geometry.json":[551],"also":[552],"writes":[553],"ordered":[555],"geometry":[557],"regions.":[561],"alone":[564,949],"use":[566],"A1's":[569,928,1041],"classes":[570,792],"types,":[573],"but":[574,594,1163],"encodes":[579],"convert":[585],"can":[586,1180],"combine":[587],"into":[591],"file,":[593],"segment":[596],"then":[597],"stops":[598],"\"Reading":[600],"class":[603,658],"mapping":[604],"\u2026":[605],"does":[606],"match":[608],"segmentation":[609],"model\".":[610],"any":[621],"case":[622],"details":[625],"data":[627,868],"synthetic.":[629,947,1166],"scraped,":[632],"third-party":[638,755],"used.":[641],"Pages:":[642],"3,990":[643],"training":[644,664,675,701,852,890,1235],"210":[646,835],"validation":[647,836],"hocrsyngen":[651,666,677,728,1259,1267],"batches,":[652],"re-prepared":[653],"so":[654,854,938,1177],"batch:":[665,676],"8698895,":[667,1260],"seed":[668,681,839,1261,1269],"5101;":[669],"1,710":[670],"/":[671,684,976,981,985,990,994,999,1006,1011,1020,1025],"90":[672],"pages.":[673,686,837],"R3d's":[674],"db3f73c":[678],"(generator":[679],"s11b-ledger-fit-v1),":[680],"7101;":[682],"2,280":[683],"120":[685],"cover":[689],"seven":[691],"templates,":[692],"565\u2013573":[693],"per":[694],"template":[695],"training.":[697],"That":[698],"gives":[699,1152],"2,181,264":[700],"pairs.":[703],"Fonts:":[704,1237],"Alef":[705,1217],"(3,000":[706],"pages)":[707],"Gveret":[709,1224,1227],"Levin":[710,1225,1228],"(1,200":[711],"pages),":[712],"both":[713],"SIL":[714],"OFL":[715],"1.1.":[716],"Credits":[717],"NOTICE.":[720],"Text:":[721],"rendered":[725],"corpora":[729],"S10b":[730,734],"(R4d's":[731],"batch)":[732],"S13a":[736],"(R3d's":[737],"batch).":[738],"original,":[742],"authored":[743],"AI":[745],"agents":[746],"(Claude":[747],"Opus":[748],"5.5":[749],"sessions)":[750],"HeOCR,":[752],"text,":[756],"dedicated":[758],"public":[761],"domain":[762],"under":[763,1062],"CC0":[764,1233],"1.0.":[765],"sees":[767],"positions":[769],"classes,":[771,931],"itself.":[775],"procedure":[777],"hyperparameters":[779],"Initialisation:":[780],"scratch.":[782],"Trainer:":[783],"-f":[787],"baselines.":[790],"passed":[794],"--config,":[796],"because":[797],"mittagessen/kraken#821.":[799],"Stopping:":[800],"early":[801],"stopping":[802],"lag":[804],"minimum":[808],"30":[810,814,893],"epochs.":[811,815],"ran":[813],"Validation:":[816],"val_spearman":[817],"(kraken's":[818],"normalised":[819],"Spearman":[820],"footrule":[821],"distance;":[822],"0":[823,1019],"perfect":[826],"order)":[827],"reached":[828],"0.0":[829],"epoch":[831],"26":[832],"Run:":[838],"42,":[840],"Apple":[842],"mps.":[843],"Not":[844],"deterministic.":[845],"record:":[847],"has":[849,1116],"command,":[853,863],"written":[858],"afterwards":[859],"config,":[864],"log":[866],"summaries.":[869],"what":[872],"evidence":[874],"supply:":[876],"defaults":[879],"logged;":[883],"precision":[885],"determinism":[887],"settings;":[888],"why":[889],"stopped":[891],"epochs;":[894],"which":[895],"epochs":[897],"26\u201329":[898],"kept":[900],"file":[901],"comes":[902],"from.":[903],"Evaluation":[904],"Testing":[905],"data,":[906,1243],"factors":[907],"metrics":[909],"those":[913,1164],"card.":[917],"None":[918],"tuned":[922],"on,":[923],"except":[924],"hocrgen":[926,1090,1312],"fixture:":[927],"design":[929],"(region-type":[930],"stamp-first":[932],"tagging)":[933],"chosen":[935],"it,":[937],"development":[942],"set.":[943],"Order":[948],"Pages":[950],"fully":[951],"over":[954],"Kendall":[960],"\u03c4":[961],"shown":[963],"where":[964],"out":[967,1112],"Batch":[970,1066],"heuristic":[972],"9":[975],"80":[977,980,982,986,989,991,995,998,1000],"(\u03c4":[978,987,996,1008,1013,1022],"0.819)":[979],"held-out":[983,1078],"12":[984],"0.865)":[988],"eval-mixed-heldout":[992,1081],"11":[993],"0.853)":[997],"fixture":[1001,1015,1091],"archive_card":[1002],"(development":[1003,1017,1092],"set)":[1004,1018,1093],"3":[1005],"25":[1007,1012,1021,1024,1026,1108],"0.917)":[1009],"15":[1010],"0.992)":[1014],"ledger":[1016,1051],"0.774)":[1023],"eval-layout-heldout,":[1028],"fresh":[1030],"batch":[1031],"ledgers":[1035],"made":[1036],"after":[1037,1128],"fixed,":[1040],"equals":[1043],"gold-order":[1045],"card":[1049],"page.":[1052],"End":[1053],"(R4d":[1056],"R3h)":[1060,1326],"CER":[1061],"hocrgen's":[1063],"primary.v1":[1064],"profile:":[1065],"blla":[1072],"0.020":[1076],"0.168":[1077],"0.010":[1079],"0.238":[1080],"0.017":[1082],"0.230":[1083],"eval-typo-heldout":[1084],"0.009":[1085],"0.291":[1086],"eval-layout-heldout":[1087],"0.124":[1088],"0.498":[1089],"0.130":[1094],"0.404":[1095],"Bias,":[1096],"risks":[1097],"limitations":[1099],"Long":[1100],"cards:":[1102],"fixture,":[1105],"still":[1111],"Each":[1115],"discordant":[1118],"pair:":[1119],"sixth":[1121],"lowest":[1123],"read":[1127],"label.":[1131],"Six-body-line":[1132],"rare":[1135],"training:":[1137],"18":[1138],"1,199":[1140],"cards.":[1141],"wrong":[1149,1154],"tag":[1151],"Tagging":[1156],"exact":[1158],"checked,":[1162],"gap.":[1169],"Only":[1170],"templates":[1172],"seen":[1174],"training,":[1176],"documents":[1179],"have":[1181],"never":[1184],"learned.":[1185],"above).":[1193],"attribution":[1196],"Licence:":[1197],"Apache-2.0.":[1198],"maintainers.":[1203],"NOTICE:":[1204],"redistributions":[1205],"must":[1206],"keep":[1207],"NOTICE":[1209],"file.":[1210],"credits:":[1212],"(Apache-2.0);":[1214],"fonts":[1216,1246],"(HaGilda":[1218],"&":[1219],"Mushon":[1220],"Zer-Aviv,":[1221],"OFL-1.1)":[1222],"(The":[1226],"Project":[1229],"Authors,":[1230],"OFL-1.1);":[1231],"text.":[1236],"contains":[1240],"font":[1242],"distributed":[1249],"it.":[1251],"Provenance":[1252],"reproduction":[1254],"Part":[1255],"Identity":[1256],"5101":[1262],"(manifest":[1263,1271],"29dffcca283770c56dedb4e53b57ccae289ec37124fd4399d968d93a3dd9295a,":[1264],"annotations":[1265,1273],"871355fdecb36fd561c1ee9832d8b6aec221685cca6ff3f13cad6b60cd6f36bf);":[1266],"db3f73c,":[1268],"7101":[1270],"e8098844b7dfb7027913d7a04e4a5d35100ab551d41bb759ec92f0f685c2ec73,":[1272],"51cab2f331607035f8a697ca3ffdcbff40241a65821fb4634dd80412b4e59c43)":[1274],"a3f2f0db1bf48d8cdd618579af42ca36710cadf81cdc4d5e89d2dc6c192e6611":[1277],"(S10b)":[1278],"e4384322e167f81252f28703206f180d1757e789c3ef7bd97484eef6978b1cf4":[1280],"(S10b":[1281],"S13a)":[1283],"Data":[1284],"preparation":[1285],"5294ca5":[1287],"prepare-kraken-data":[1288],"--task":[1289],"segmentation,":[1290],"name":[1293],"rotrain,":[1302],"2026-11-04":[1303],"(heocr_engine_training_run.v1),":[1307],"deposit":[1310],"Scoring":[1311],"2d3efb0":[1313],"c8fb74e;":[1315],"4e09858":[1317],"660e5a4":[1319],"(order),":[1320],"55b0f7d":[1321],"(end":[1322],"Contact":[1327],"Report":[1328],"problems":[1329],"questions":[1331],"issues":[1333],"HeOCR/heocr-engine.":[1335],"Citation":[1336],"Cite":[1337],"DOI,":[1343],"10.5281/zenodo.24023072.":[1344],"Never":[1345],"cite":[1346],"concept":[1348],"DOI.":[1349]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-11-07T00:00:00"}