{"id":"https://openalex.org/W7166683551","doi":"https://doi.org/10.48550/arxiv.2606.28780","title":"Multimodal Graph RAG for Long-range Visually Rich Document Understanding","display_name":"Multimodal Graph RAG for Long-range Visually Rich Document Understanding","publication_year":2026,"publication_date":"2026-06-27","ids":{"openalex":"https://openalex.org/W7166683551","doi":"https://doi.org/10.48550/arxiv.2606.28780"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.28780","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.28780","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.28780","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139679389","display_name":"Yi-Cheng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yi-Cheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139672377","display_name":"Chu-Song Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Chu-Song","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":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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9606999754905701,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9606999754905701,"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/T10028","display_name":"Topic Modeling","score":0.009399999864399433,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.0071000000461936,"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/context","display_name":"Context (archaeology)","score":0.4603999853134155},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4440999925136566},{"id":"https://openalex.org/keywords/knowledge-graph","display_name":"Knowledge graph","score":0.3781999945640564},{"id":"https://openalex.org/keywords/comprehension","display_name":"Comprehension","score":0.3776000142097473},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.35920000076293945},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.3400999903678894}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8319000005722046},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46639999747276306},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4603999853134155},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4440999925136566},{"id":"https://openalex.org/C2987255567","wikidata":"https://www.wikidata.org/wiki/Q33002955","display_name":"Knowledge graph","level":2,"score":0.3781999945640564},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.3776000142097473},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.35920000076293945},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3450999855995178},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3443000018596649},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3400999903678894},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.33390000462532043},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.27250000834465027},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.26980000734329224},{"id":"https://openalex.org/C135641252","wikidata":"https://www.wikidata.org/wiki/Q738567","display_name":"Multimodal interaction","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2549000084400177}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.28780","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.28780","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.28780","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.28780","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":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"large":[1],"language":[2,82],"models":[3],"(MLLMs)":[4],"are":[5],"widely":[6],"applied":[7],"to":[8,108,145],"visual":[9,43],"document":[10,66],"understanding.":[11],"However,":[12,72],"comprehending":[13],"long":[14],"documents":[15,95],"remains":[16],"an":[17,69],"issue":[18],"by":[19,34],"the":[20,42,81,85,117],"limited":[21],"context":[22],"window.":[23],"Though":[24],"recent":[25],"multimodal":[26,89,104],"retrieval-augmented":[27],"generation":[28],"(MMRAG)":[29],"can":[30,67],"address":[31],"this":[32,99,110,151],"challenge":[33,144],"retrieving":[35],"relevant":[36],"pages.":[37],"It":[38],"still":[39],"struggles":[40],"with":[41,55],"question":[44],"answering":[45],"(VQA)":[46],"requiring":[47],"holistic":[48],"comprehension":[49],"of":[50,64,88,134],"a":[51,65,103,142,156],"document.":[52],"To":[53,149],"cope":[54],"this,":[56],"knowledge":[57,63],"graph":[58],"(KG)":[59],"that":[60,177],"summarizes":[61],"global":[62,171],"provide":[68],"effective":[70,146],"solution.":[71],"most":[73],"existing":[74,181],"LLM-based":[75,113],"KG":[76,114],"construction":[77],"methods":[78,115],"handle":[79],"only":[80],"modality,":[83],"leaving":[84],"automatic":[86],"creation":[87],"KGs":[90],"(MMKGs)":[91],"for":[92,137,170],"visually":[93],"rich":[94],"largely":[96],"unexplored.":[97],"In":[98],"paper,":[100],"we":[101,153],"introduce":[102,155],"graph-based":[105],"RAG":[106],"approach":[107,179],"tackle":[109],"problem.":[111],"Existing":[112],"evaluate":[116],"QA":[118],"performance":[119],"relying":[120],"on":[121,186],"indirect":[122],"evidence":[123],"such":[124],"as":[125],"comprehensiveness,":[126],"diversity,":[127],"empowerment,":[128],"and":[129,166,190],"so":[130],"on.":[131],"The":[132],"lack":[133],"annotated":[135],"datasets":[136],"comprehensive":[138],"document-level":[139,172],"VQA":[140],"poses":[141],"significant":[143],"model":[147],"evaluation.":[148],"overcome":[150],"limitation,":[152],"also":[154],"new":[157],"benchmark,":[158],"DLVQA":[159],"(document-level":[160],"VQA),":[161],"which":[162],"provides":[163],"reference":[164],"summaries":[165],"corresponding":[167],"supporting":[168],"facts":[169],"questions.":[173],"Experimental":[174],"results":[175],"show":[176],"our":[178],"outperforms":[180],"MMRAG":[182],"or":[183],"KG-based":[184],"approaches":[185],"multi-hop":[187],"QA/VQA":[188],"benchmarks":[189],"DLVQA.":[191]},"counts_by_year":[],"updated_date":"2026-07-01T06:29:00.853634","created_date":"2026-07-01T00:00:00"}