{"id":"https://openalex.org/W7140797831","doi":"https://doi.org/10.48550/arxiv.2603.23885","title":"Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training","display_name":"Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training","publication_year":2026,"publication_date":"2026-03-25","ids":{"openalex":"https://openalex.org/W7140797831","doi":"https://doi.org/10.48550/arxiv.2603.23885"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.23885","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23885","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.23885","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058261095","display_name":"Gengluo Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Gengluo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Lyu, Pengyuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lyu, Pengyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang, Chengquan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Chengquan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122369826","display_name":"Huawen Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Huawen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130714088","display_name":"Yaping Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wan, Xingyu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wan, Xingyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zeng, Gangyan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Gangyan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102652743","display_name":"Xingyu Wan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046229717","display_name":"Gangyan Zeng","orcid":"https://orcid.org/0000-0003-2696-8549"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Can","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130643804","display_name":"Han Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":10,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5008000135421753,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5008000135421753,"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/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.2134999930858612,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.10599999874830246,"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/robustness","display_name":"Robustness (evolution)","score":0.7519000172615051},{"id":"https://openalex.org/keywords/parsing","display_name":"Parsing","score":0.7268000245094299},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.5253999829292297},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.44670000672340393},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.38280001282691956},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.34369999170303345}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8162000179290771},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7519000172615051},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.7268000245094299},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.5253999829292297},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5080999732017517},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.44670000672340393},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.38280001282691956},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37860000133514404},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.34369999170303345},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.34060001373291016},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.32850000262260437},{"id":"https://openalex.org/C72773152","wikidata":"https://www.wikidata.org/wiki/Q5287629","display_name":"Document layout analysis","level":3,"score":0.3257000148296356},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.30570000410079956},{"id":"https://openalex.org/C35639132","wikidata":"https://www.wikidata.org/wiki/Q7452468","display_name":"Sequence labeling","level":3,"score":0.2904999852180481},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.2856000065803528},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.2847999930381775},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2777999937534332},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2574999928474426},{"id":"https://openalex.org/C2988504005","wikidata":"https://www.wikidata.org/wiki/Q379942","display_name":"Document image processing","level":4,"score":0.25699999928474426}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.23885","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23885","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.23885","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.23885","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Document":[0],"parsing":[1,63],"has":[2],"recently":[3],"advanced":[4],"with":[5,103],"multimodal":[6],"large":[7],"language":[8],"models":[9],"(MLLMs)":[10],"that":[11],"directly":[12],"map":[13],"document":[14,85,105,174],"images":[15],"to":[16,54,118,169],"structured":[17],"outputs.":[18],"Traditional":[19],"cascaded":[20],"pipelines":[21],"depend":[22],"on":[23],"precise":[24],"layout":[25,101],"analysis":[26],"and":[27,47,65,115,122,149,154,163],"often":[28],"fail":[29],"under":[30],"casually":[31],"captured":[32,134,156],"or":[33],"non-standard":[34],"conditions.":[35],"Although":[36],"end-to-end":[37,62,84,97],"approaches":[38],"mitigate":[39],"this":[40],"dependency,":[41],"they":[42],"still":[43],"exhibit":[44],"repetitive,":[45],"hallucinated,":[46],"structurally":[48,94],"inconsistent":[49],"predictions":[50],"-":[51],"primarily":[52],"due":[53],"the":[55,66],"scarcity":[56],"of":[57,68],"large-scale,":[58,93],"high-quality":[59],"full-page":[60,96],"(document-level)":[61],"data":[64,160],"lack":[67],"structure-aware":[69],"training":[70],"strategies.":[71],"To":[72],"address":[73],"these":[74],"challenges,":[75],"we":[76],"propose":[77],"a":[78,108,129,141],"data-training":[79],"co-design":[80],"framework":[81],"for":[82,136],"robust":[83],"parsing.":[86],"A":[87],"Realistic":[88],"Scene":[89],"Synthesis":[90],"strategy":[91],"constructs":[92],"diverse":[95],"supervision":[98],"by":[99],"composing":[100],"templates":[102],"rich":[104],"elements,":[106],"while":[107],"Document-Aware":[109],"Training":[110],"Recipe":[111],"introduces":[112],"progressive":[113],"learning":[114],"structure-token":[116],"optimization":[117],"enhance":[119],"structural":[120],"fidelity":[121],"decoding":[123],"stability.":[124],"We":[125],"further":[126],"build":[127],"Wild-OmniDocBench,":[128],"benchmark":[130],"derived":[131],"from":[132],"real-world":[133,155],"documents":[135],"robustness":[137,150],"evaluation.":[138],"Integrated":[139],"into":[140],"1B-parameter":[142],"MLLM,":[143],"our":[144],"method":[145],"achieves":[146],"superior":[147],"accuracy":[148],"across":[151],"both":[152],"scanned/digital":[153],"scenarios.":[157],"All":[158],"models,":[159],"synthesis":[161],"pipelines,":[162],"benchmarks":[164],"will":[165],"be":[166],"publicly":[167],"released":[168],"advance":[170],"future":[171],"research":[172],"in":[173],"understanding.":[175]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-27T00:00:00"}