{"id":"https://openalex.org/W4399694974","doi":"https://doi.org/10.48550/arxiv.2406.08487","title":"Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal Models","display_name":"Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal Models","publication_year":2024,"publication_date":"2024-06-12","ids":{"openalex":"https://openalex.org/W4399694974","doi":"https://doi.org/10.48550/arxiv.2406.08487"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2406.08487","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.08487","pdf_url":"https://arxiv.org/pdf/2406.08487","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2406.08487","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100376940","display_name":"Yifan Zhang","orcid":"https://orcid.org/0000-0002-6227-0183"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhang, Yi-Fan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048346353","display_name":"Qingsong Wen","orcid":"https://orcid.org/0000-0003-4516-2524"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wen, Qingsong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014172220","display_name":"Chaoyou Fu","orcid":"https://orcid.org/0000-0002-0079-7668"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Chaoyou","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084186029","display_name":"X. Wang","orcid":"https://orcid.org/0000-0003-4954-9216"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xue","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100599317","display_name":"Zhang Zhang","orcid":"https://orcid.org/0000-0002-9331-9699"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wang, Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Liang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5101873437","display_name":"Rong Jin","orcid":"https://orcid.org/0000-0003-1789-236X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin, Rong","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5100376940"],"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":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13734","display_name":"Advanced Computational Techniques and Applications","score":0.597000002861023,"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/T13734","display_name":"Advanced Computational Techniques and Applications","score":0.597000002861023,"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/computer-science","display_name":"Computer science","score":0.41981056332588196},{"id":"https://openalex.org/keywords/resolution","display_name":"Resolution (logic)","score":0.4181596636772156},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.18746554851531982}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.41981056332588196},{"id":"https://openalex.org/C138268822","wikidata":"https://www.wikidata.org/wiki/Q1051925","display_name":"Resolution (logic)","level":2,"score":0.4181596636772156},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.18746554851531982}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2406.08487","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.08487","pdf_url":"https://arxiv.org/pdf/2406.08487","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2406.08487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2406.08487","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:2406.08487","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.08487","pdf_url":"https://arxiv.org/pdf/2406.08487","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4399694974.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W2382290278","https://openalex.org/W4395014643"],"abstract_inverted_index":{"Seeing":[0],"clearly":[1],"with":[2,43,235,254,272],"high":[3,236],"resolution":[4,31,56,64],"is":[5,169],"a":[6,29,97,117,160,167,186,232],"foundation":[7],"of":[8,38,78,119,198,244,262],"Large":[9],"Multimodal":[10],"Models":[11],"(LMMs),":[12],"which":[13],"has":[14],"been":[15],"proven":[16],"to":[17,53,134,142,182],"be":[18,156],"vital":[19],"for":[20,151,215,238],"visual":[21],"perception":[22],"and":[23,40,74,95,203,225,260],"reasoning.":[24],"Existing":[25],"works":[26],"usually":[27],"employ":[28],"straightforward":[30],"upscaling":[32],"method,":[33,251],"where":[34,172],"the":[35,44,47,54,58,76,84,93,113,123,146,152,191,199,242,245],"image":[36,49,80,144,179,239,258],"consists":[37],"global":[39,85,114,200,224,263],"local":[41,67,79,135,178,204,226,246],"branches,":[42],"latter":[45],"being":[46],"sliced":[48],"patches":[50],"but":[51,175],"resized":[52],"same":[55],"as":[57,100,102,194],"former.":[59],"This":[60],"means":[61],"that":[62,125],"higher":[63],"requires":[65],"more":[66,176],"patches,":[68,136],"resulting":[69],"in":[70,190],"exorbitant":[71],"computational":[72],"expenses,":[73],"meanwhile,":[75],"dominance":[77],"tokens":[81,149,180],"may":[82],"diminish":[83],"context.":[86],"In":[87],"this":[88],"paper,":[89],"we":[90,108,229],"dive":[91],"into":[92],"problems":[94],"propose":[96],"new":[98],"framework":[99],"well":[101],"an":[103,216],"elaborate":[104],"optimization":[105],"strategy.":[106],"Specifically,":[107],"extract":[109],"contextual":[110],"information":[111],"from":[112],"view":[115],"using":[116],"mixture":[118],"adapters,":[120],"based":[121],"on":[122],"observation":[124],"different":[126,130],"adapters":[127],"excel":[128],"at":[129],"tasks.":[131],"With":[132],"regard":[133],"learnable":[137],"query":[138],"embeddings":[139],"are":[140],"introduced":[141],"reduce":[143],"tokens,":[145],"most":[147],"important":[148],"accounting":[150],"user":[153],"question":[154],"will":[155],"further":[157],"selected":[158],"by":[159],"similarity-based":[161],"selector.":[162],"Our":[163],"empirical":[164],"results":[165],"demonstrate":[166],"`less":[168],"more'":[170],"pattern,":[171],"\\textit{utilizing":[173],"fewer":[174],"informative":[177],"leads":[181],"improved":[183],"performance}.":[184],"Besides,":[185],"significant":[187],"challenge":[188],"lies":[189],"training":[192,197,218,243,276],"strategy,":[193],"simultaneous":[195],"end-to-end":[196],"mining":[201],"block":[202,206],"compression":[205,247],"does":[207],"not":[208],"yield":[209],"optimal":[210],"results.":[211],"We":[212],"thus":[213],"advocate":[214],"alternating":[217],"way,":[219],"ensuring":[220],"balanced":[221],"learning":[222],"between":[223],"aspects.":[227],"Finally,":[228],"also":[230],"introduce":[231],"challenging":[233],"dataset":[234],"requirements":[237],"detail,":[240],"enhancing":[241],"layer.":[248],"The":[249],"proposed":[250],"termed":[252],"LMM":[253],"Sophisticated":[255],"Tasks,":[256],"Local":[257],"compression,":[259],"Mixture":[261],"Experts":[264],"(SliME),":[265],"achieves":[266],"leading":[267],"performance":[268],"across":[269],"various":[270],"benchmarks":[271],"only":[273],"2":[274],"million":[275],"data.":[277]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}