{"id":"https://openalex.org/W4417296427","doi":"https://doi.org/10.48550/arxiv.2512.10619","title":"DOCR-Inspector: Fine-Grained and Automated Evaluation of Document Parsing with VLM","display_name":"DOCR-Inspector: Fine-Grained and Automated Evaluation of Document Parsing with VLM","publication_year":2025,"publication_date":"2025-12-11","ids":{"openalex":"https://openalex.org/W4417296427","doi":"https://doi.org/10.48550/arxiv.2512.10619"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2512.10619","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.10619","pdf_url":"https://arxiv.org/pdf/2512.10619","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2512.10619","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022441996","display_name":"Qintong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhang, Qintong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073244430","display_name":"Junyuan Zhang","orcid":"https://orcid.org/0000-0003-1593-4888"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Junyuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Ren, Zhifei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ren, Zhifei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104670548","display_name":"Linke Ouyang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ouyang, Linke","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102635197","display_name":"Zichen Wen","orcid":"https://orcid.org/0009-0002-6157-5898"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wen, Zichen","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085823891","display_name":"Junbo Niu","orcid":"https://orcid.org/0000-0002-2135-6853"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Niu, Junbo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115587888","display_name":"Yuan Qu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qu, Yuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101958711","display_name":"Bin Wang","orcid":"https://orcid.org/0000-0002-9607-1158"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Bin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102463289","display_name":"Ka-Ho Chow","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chow, Ka-Ho","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101993018","display_name":"Cong He","orcid":"https://orcid.org/0000-0003-1088-8915"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Conghui","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5008772211","display_name":"Wentao Zhang","orcid":"https://orcid.org/0000-0002-7532-5550"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Wentao","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":11,"corresponding_author_ids":["https://openalex.org/A5022441996"],"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/T10181","display_name":"Natural Language Processing Techniques","score":0.4975999891757965,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10181","display_name":"Natural Language Processing Techniques","score":0.4975999891757965,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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":0.15860000252723694,"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":0.055799998342990875,"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/parsing","display_name":"Parsing","score":0.8317000269889832},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.545199990272522},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5128999948501587},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5019999742507935},{"id":"https://openalex.org/keywords/bottom-up-parsing","display_name":"Bottom-up parsing","score":0.47600001096725464},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4235999882221222},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.4115999937057495}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8646000027656555},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.8317000269889832},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6172000169754028},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5551000237464905},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.545199990272522},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5128999948501587},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5019999742507935},{"id":"https://openalex.org/C60690694","wikidata":"https://www.wikidata.org/wiki/Q894902","display_name":"Bottom-up parsing","level":4,"score":0.47600001096725464},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4235999882221222},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4115999937057495},{"id":"https://openalex.org/C42560504","wikidata":"https://www.wikidata.org/wiki/Q15419395","display_name":"Top-down parsing","level":3,"score":0.3772999942302704},{"id":"https://openalex.org/C147547768","wikidata":"https://www.wikidata.org/wiki/Q3113342","display_name":"S-attributed grammar","level":3,"score":0.37400001287460327},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37119999527931213},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.3668000102043152},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3652999997138977},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.31690001487731934},{"id":"https://openalex.org/C2779308522","wikidata":"https://www.wikidata.org/wiki/Q843958","display_name":"Digitization","level":2,"score":0.30799999833106995},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.28119999170303345},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.2671999931335449},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2644999921321869}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2512.10619","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.10619","pdf_url":"https://arxiv.org/pdf/2512.10619","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2512.10619","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.10619","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":"pmh:oai:arXiv.org:2512.10619","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.10619","pdf_url":"https://arxiv.org/pdf/2512.10619","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"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":[1,34,97,111,168,183,216,231],"aims":[2],"to":[3,58,135,162],"transform":[4],"unstructured":[5],"PDF":[6],"images":[7],"into":[8],"semi-structured":[9],"data,":[10],"facilitating":[11],"the":[12,44,100,158,164],"digitization":[13],"and":[14,62,93,117,126,141,156,225,236],"utilization":[15],"of":[16,137,167,179],"information":[17],"in":[18,35,82,99],"diverse":[19],"domains.":[20],"While":[21],"vision":[22],"language":[23],"models":[24,194],"(VLMs)":[25],"have":[26],"significantly":[27],"advanced":[28],"this":[29,104,149,189],"task,":[30],"achieving":[31],"reliable,":[32],"high-quality":[33],"real-world":[36,66,181],"scenarios":[37],"remains":[38],"challenging.":[39],"Common":[40],"practice":[41],"often":[42],"selects":[43],"top-performing":[45],"model":[46,60],"on":[47],"standard":[48],"benchmarks.":[49],"However,":[50],"these":[51],"benchmarks":[52],"may":[53],"carry":[54],"dataset-specific":[55],"biases,":[56],"leading":[57,202],"inconsistent":[59],"rankings":[61],"limited":[63],"correlation":[64],"with":[65,106,185],"performance.":[67],"Moreover,":[68],"benchmark":[69],"metrics":[70],"typically":[71],"provide":[72,212],"only":[73],"overall":[74],"scores,":[75],"which":[76,108],"can":[77,90],"obscure":[78],"distinct":[79],"error":[80,115],"patterns":[81],"output.":[83],"This":[84],"raises":[85],"a":[86,123,143,177,222,226],"key":[87],"challenge:":[88],"how":[89],"we":[91,151,174],"reliably":[92],"comprehensively":[94],"assess":[95],"document":[96,110,124,182,230],"quality":[98,145,169,210],"wild?":[101],"We":[102],"address":[103],"problem":[105],"DOCR-Inspector,":[107],"formalizes":[109],"assessment":[112],"as":[113,199,201],"fine-grained":[114],"detection":[116],"analysis.":[118],"Leveraging":[119],"VLM-as-a-Judge,":[120],"DOCR-Inspector":[121,220],"analyzes":[122],"image":[125],"its":[127,209],"parsed":[128],"output,":[129],"identifies":[130],"all":[131],"errors,":[132],"assigns":[133],"them":[134],"one":[136],"28":[138],"predefined":[139],"types,":[140],"produces":[142],"comprehensive":[144],"assessment.":[146,170],"To":[147],"enable":[148,163],"capability,":[150],"construct":[152],"DOCRcase-200K":[153],"for":[154,215,228],"training":[155],"propose":[157],"Chain-of-Checklist":[159],"reasoning":[160],"paradigm":[161],"hierarchical":[165],"structure":[166],"For":[171],"empirical":[172],"validation,":[173],"introduce":[175],"DOCRcaseBench,":[176],"set":[178],"882":[180],"cases":[184],"manual":[186],"annotations.":[187],"On":[188],"benchmark,":[190],"DOCR-Inspector-7B":[191],"outperforms":[192],"commercial":[193],"like":[195],"Gemini":[196],"2.5":[197],"Pro,":[198],"well":[200],"open-source":[203],"models.":[204],"Further":[205],"experiments":[206],"demonstrate":[207],"that":[208],"assessments":[211],"valuable":[213],"guidance":[214],"results":[217],"refinement,":[218],"making":[219],"both":[221],"practical":[223],"evaluator":[224],"driver":[227],"advancing":[229],"systems":[232],"at":[233],"scale.":[234],"Model":[235],"code":[237],"are":[238],"released":[239],"at:":[240],"https://github.com/ZZZZZQT/DOCR-Inspector.":[241]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-12-13T00:00:00"}