{"id":"https://openalex.org/W7140112893","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.226","title":"Efficient Table Retrieval and Understanding with Multimodal Large Language Models","display_name":"Efficient Table Retrieval and Understanding with Multimodal Large Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140112893","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.226"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-eacl.226","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.226","pdf_url":"https://aclanthology.org/2026.findings-eacl.226.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-eacl.226.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Zhuoyan Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuoyan Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Haoyang Fang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haoyang Fang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Boran Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Boran Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Bonan Min","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bonan Min","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Bernie Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bernie Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Cuixiong Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cuixiong Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Shuai Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shuai Zhang","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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4327","last_page":"4340"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.357699990272522,"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":0.357699990272522,"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/T11719","display_name":"Data Quality and Management","score":0.17270000278949738,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T13083","display_name":"Advanced Text Analysis Techniques","score":0.05640000104904175,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/table","display_name":"Table (database)","score":0.412200003862381},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.33469998836517334},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3176000118255615},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.29829999804496765},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.28139999508857727}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7095000147819519},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4805999994277954},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4781999886035919},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.412200003862381},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3176000118255615},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3131999969482422},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.28139999508857727},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.25619998574256897}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-eacl.226","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.226","pdf_url":"https://aclanthology.org/2026.findings-eacl.226.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-eacl.226","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.226","pdf_url":"https://aclanthology.org/2026.findings-eacl.226.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.49453434348106384,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140112893.pdf","grobid_xml":"https://content.openalex.org/works/W7140112893.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Tabular":[0],"data":[1],"is":[2,60],"frequently":[3],"captured":[4],"in":[5,42,51,167,172],"image":[6],"form":[7],"across":[8,149],"a":[9,63,87,138,176],"wide":[10],"range":[11],"of":[12,98,119],"real-world":[13],"scenarios":[14],"such":[15],"as":[16,32],"financial":[17],"reports,":[18],"handwritten":[19],"records,":[20],"and":[21,37,69,122,145,170],"document":[22],"scans.These":[23],"visual":[24,38],"representations":[25],"pose":[26],"unique":[27,154],"challenges":[28],"for":[29,132,179],"machine":[30],"understanding,":[31,53],"they":[33,54],"combine":[34],"both":[35],"structural":[36],"complexities.While":[39],"recent":[40],"advances":[41],"Multimodal":[43],"Large":[44],"Language":[45],"Models":[46],"(MLLMs)":[47],"show":[48],"promising":[49],"results":[50],"table":[52,59,99,181],"typically":[55],"assume":[56],"the":[57,129],"relevant":[58,72],"readily":[61],"available.However,":[62],"more":[64],"practical":[65,177],"scenario":[66],"involves":[67],"identifying":[68],"reasoning":[70],"over":[71,95,128],"tables":[73,105,131],"from":[74],"largescale":[75],"collections":[76,97],"to":[77,92,115,126],"answer":[78,93,133,173],"user":[79],"queries.To":[80],"address":[81],"this":[82],"gap,":[83],"we":[84,156],"propose":[85],"TabRAG,":[86],"framework":[88,160],"that":[89,158],"enables":[90],"MLLMs":[91,114,125],"queries":[94],"large":[96],"images.Our":[100],"approach":[101],"first":[102],"retrieves":[103],"candidate":[104],"using":[106],"jointly":[107],"trained":[108],"visual-text":[109],"foundation":[110],"models,":[111],"then":[112],"leverages":[113],"perform":[116],"fine-grained":[117],"reranking":[118],"these":[120],"candidates,":[121],"finally":[123],"employs":[124],"reason":[127],"selected":[130],"generation.Through":[134],"extensive":[135],"experiments":[136],"on":[137],"newly":[139],"constructed":[140],"dataset":[141],"comprising":[142],"88,161":[143],"training":[144],"9,819":[146],"testing":[147],"samples":[148],"8":[150],"benchmarks":[151],"with":[152],"48,504":[153],"tables,":[155],"demonstrate":[157],"our":[159],"significantly":[161],"outperforms":[162],"existing":[163],"methods":[164],"by":[165],"7.0%":[166],"retrieval":[168],"recall":[169],"6.1%":[171],"accuracy,":[174],"offering":[175],"solution":[178],"realworld":[180],"understanding":[182],"tasks.":[183]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-02-12T00:00:00"}