{"id":"https://openalex.org/W4403791048","doi":"https://doi.org/10.1145/3664647.3680750","title":"CREAM: Coarse-to-Fine Retrieval and Multi-modal Efficient Tuning for Document VQA","display_name":"CREAM: Coarse-to-Fine Retrieval and Multi-modal Efficient Tuning for Document VQA","publication_year":2024,"publication_date":"2024-10-26","ids":{"openalex":"https://openalex.org/W4403791048","doi":"https://doi.org/10.1145/3664647.3680750"},"language":"en","primary_location":{"id":"doi:10.1145/3664647.3680750","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3680750","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102878671","display_name":"Jinxu Zhang","orcid":"https://orcid.org/0009-0000-9876-1454"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinxu Zhang","raw_affiliation_strings":["Harbin Institute of Technology, Harbin, Heilongjiang, China"],"raw_orcid":"https://orcid.org/0009-0000-9876-1454","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology, Harbin, Heilongjiang, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101188257","display_name":"Yongqi Yu","orcid":"https://orcid.org/0009-0005-0674-788X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yongqi Yu","raw_affiliation_strings":["Harbin Institute of Technology, Harbin, Heilongjiang, China"],"raw_orcid":"https://orcid.org/0009-0005-0674-788X","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology, Harbin, Heilongjiang, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100433723","display_name":"Yu Zhang","orcid":"https://orcid.org/0000-0003-3090-7431"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Zhang","raw_affiliation_strings":["Harbin Institute of Technology, Harbin, Heilongjiang, China"],"raw_orcid":"https://orcid.org/0000-0003-3090-7431","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology, Harbin, Heilongjiang, China","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I204983213"],"apc_list":null,"apc_paid":null,"fwci":1.0358,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.82727438,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"925","last_page":"934"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9994999766349792,"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/T10028","display_name":"Topic Modeling","score":0.9977999925613403,"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/modal","display_name":"Modal","score":0.7429779171943665},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6747823357582092},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.5819194316864014},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3690541386604309},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.10698890686035156},{"id":"https://openalex.org/keywords/composite-material","display_name":"Composite material","score":0.05725771188735962}],"concepts":[{"id":"https://openalex.org/C71139939","wikidata":"https://www.wikidata.org/wiki/Q910194","display_name":"Modal","level":2,"score":0.7429779171943665},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6747823357582092},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.5819194316864014},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3690541386604309},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.10698890686035156},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.05725771188735962}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3664647.3680750","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3680750","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W1956340063","https://openalex.org/W2988326850","https://openalex.org/W2997154779","https://openalex.org/W2999905431","https://openalex.org/W3027879771","https://openalex.org/W3104953317","https://openalex.org/W3120043490","https://openalex.org/W3173585224","https://openalex.org/W4224308101","https://openalex.org/W4304013646","https://openalex.org/W4312233877","https://openalex.org/W4385237494","https://openalex.org/W4386065837","https://openalex.org/W4390873989","https://openalex.org/W4393154852","https://openalex.org/W6810081322"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Document":[0],"Visual":[1],"Question":[2],"Answering":[3],"(DVQA)":[4],"involves":[5,112],"responding":[6],"to":[7,20,44,70,124,131,150,166,181],"queries":[8],"based":[9],"on":[10,39,61],"the":[11,35,45,78,99,102,109,126,132,145,153,160],"contents":[12],"of":[13,47,80,115,156],"document":[14,68,142,183,197],"images.":[15],"Existing":[16],"works":[17],"are":[18],"confined":[19],"locating":[21],"information":[22,69,155],"within":[23],"a":[24,89,120],"single":[25],"page":[26],"and":[27,64,92,104,117,171],"lack":[28],"support":[29],"for":[30,140],"cross-page":[31],"question-and-answer":[32],"interactions.":[33],"Furthermore,":[34],"token":[36],"length":[37],"limitation":[38],"model":[40,123,164],"inputs":[41],"can":[42],"lead":[43],"truncation":[46],"answer-relevant":[48],"segments.":[49],"In":[50],"this":[51,73],"study,":[52],"we":[53,86,175],"present":[54],"CREAM,":[55],"an":[56,136],"innovative":[57],"methodology":[58],"that":[59,189],"focuses":[60],"high-performance":[62],"retrieval":[63,91],"integrates":[65],"relevant":[66,130],"multimodal":[67,161],"effectively":[71,151],"address":[72],"critical":[74],"issue.":[75],"To":[76],"overcome":[77],"limitations":[79],"current":[81],"text":[82,105,127],"embedding":[83],"similarity":[84,100],"methods,":[85],"first":[87],"employ":[88],"coarse-to-fine":[90],"ranking":[93],"approach.":[94],"The":[95],"coarse":[96],"phase":[97,111],"calculates":[98],"between":[101],"query":[103],"chunk":[106],"embeddings,":[107],"while":[108],"fine":[110],"multiple":[113],"rounds":[114],"grouping":[116],"ordering":[118],"with":[119],"large":[121,162],"language":[122,163],"identify":[125],"chunks":[128],"most":[129],"query.":[133],"Subsequently,":[134],"integrating":[135],"attention":[137],"pooling":[138],"mechanism":[139],"multi-page":[141,157,172],"images":[143],"into":[144],"vision":[146],"encoder":[147],"allows":[148],"us":[149],"merge":[152],"visual":[154,184],"documents,":[158],"enabling":[159],"(MLLM)":[165],"simultaneously":[167],"process":[168],"both":[169],"single-page":[170],"documents.":[173],"Finally,":[174],"apply":[176],"various":[177,196],"parameter-efficient":[178],"tuning":[179],"methods":[180],"enhance":[182],"question-answering":[185],"performance.":[186],"Experiments":[187],"demonstrate":[188],"our":[190],"approach":[191],"secures":[192],"state-of-the-art":[193],"results":[194],"across":[195],"datasets.":[198]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-08-08T07:41:36.138363","created_date":"2025-10-10T00:00:00"}