{"id":"https://openalex.org/W4414360137","doi":"https://doi.org/10.24963/ijcai.2025/123","title":"Incorporating Visual Experts to Resolve the Information Loss in Multimodal Large Language Models","display_name":"Incorporating Visual Experts to Resolve the Information Loss in Multimodal Large Language Models","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4414360137","doi":"https://doi.org/10.24963/ijcai.2025/123"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2025/123","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/123","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"article","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/A5110117463","display_name":"Xin He","orcid":"https://orcid.org/0000-0003-0755-0964"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Xin He","raw_affiliation_strings":["Huawei Inc"],"affiliations":[{"raw_affiliation_string":"Huawei Inc","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050660610","display_name":"Longhui Wei","orcid":"https://orcid.org/0000-0001-6916-3009"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Longhui Wei","raw_affiliation_strings":["Huawei Inc"],"affiliations":[{"raw_affiliation_string":"Huawei Inc","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006960979","display_name":"Linhai Xie","orcid":"https://orcid.org/0000-0001-8593-2277"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lingxi Xie","raw_affiliation_strings":["Huawei Inc"],"affiliations":[{"raw_affiliation_string":"Huawei Inc","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5111504451","display_name":"Qi Tian","orcid":"https://orcid.org/0009-0003-2676-5300"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qi Tian","raw_affiliation_strings":["Huawei Inc"],"affiliations":[{"raw_affiliation_string":"Huawei Inc","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5110117463"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13981527,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1098","last_page":"1106"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9718000292778015,"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/T10028","display_name":"Topic Modeling","score":0.9718000292778015,"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":0.9695000052452087,"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":0.9318000078201294,"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/automatic-summarization","display_name":"Automatic summarization","score":0.7483999729156494},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6251000165939331},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.5738000273704529},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.550599992275238},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.4521999955177307},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.4323999881744385},{"id":"https://openalex.org/keywords/visual-language","display_name":"Visual language","score":0.4074000120162964}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7745000123977661},{"id":"https://openalex.org/C170858558","wikidata":"https://www.wikidata.org/wiki/Q1394144","display_name":"Automatic summarization","level":2,"score":0.7483999729156494},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6251000165939331},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.5738000273704529},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.550599992275238},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5241000056266785},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.4521999955177307},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.4323999881744385},{"id":"https://openalex.org/C2780878386","wikidata":"https://www.wikidata.org/wiki/Q1659648","display_name":"Visual language","level":2,"score":0.4074000120162964},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.4018999934196472},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.350600004196167},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.349700003862381},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.33959999680519104},{"id":"https://openalex.org/C2988416141","wikidata":"https://www.wikidata.org/wiki/Q6031139","display_name":"Information loss","level":2,"score":0.2838999927043915},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2775999903678894},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C33326189","wikidata":"https://www.wikidata.org/wiki/Q17092450","display_name":"Information integration","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.25609999895095825},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2549000084400177}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2025/123","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/123","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":null,"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],"Models":[3],"(MLLMs)":[4],"are":[5,56],"experiencing":[6],"rapid":[7],"growth,":[8],"yielding":[9],"a":[10,99,108,129],"plethora":[11],"of":[12,36,60,96,133,142,154],"novel":[13,109],"works":[14],"recently.":[15],"The":[16],"prevailing":[17],"trend":[18],"involves":[19],"adopting":[20],"data-driven":[21],"methodologies,":[22],"wherein":[23],"diverse":[24],"instruction-following":[25],"datasets":[26],"were":[27],"collected.":[28],"However,":[29],"these":[30,54],"approaches":[31],"always":[32],"face":[33],"the":[34,67,78,92,120,152],"challenge":[35],"limited":[37],"visual":[38,49,93,117,134,147,155],"perception":[39,94,148],"capabilities,":[40],"as":[41],"they":[42,63],"solely":[43],"utilizing":[44],"CLIP-like":[45],"encoders":[46,55,114],"to":[47,90,127],"extract":[48],"information":[50,68],"from":[51],"inputs.":[52,135],"Though":[53],"pre-trained":[57],"on":[58],"billions":[59],"image-text":[61],"pairs,":[62],"still":[64],"grapple":[65],"with":[66],"loss":[69],"dilemma,":[70],"given":[71],"that":[72,111],"textual":[73],"captions":[74],"only":[75],"partially":[76],"capture":[77],"contents":[79],"depicted":[80],"in":[81],"images.":[82],"To":[83],"address":[84],"this":[85,87,105],"limitation,":[86],"paper":[88],"proposes":[89],"improve":[91],"ability":[95],"MLLMs":[97,121],"through":[98,151],"mixture-of-experts":[100],"knowledge":[101],"enhancement":[102],"mechanism.":[103],"Specifically,":[104],"work":[106],"introduces":[107],"method":[110],"incorporates":[112],"multi-task":[113],"and":[115,123],"existing":[116],"tools":[118],"into":[119],"training":[122],"inference":[124],"pipeline,":[125],"aiming":[126],"provide":[128],"more":[130],"comprehensive":[131],"summarization":[132],"Extensive":[136],"experiments":[137],"have":[138],"evaluated":[139],"its":[140],"effectiveness":[141],"advancing":[143],"MLLMs,":[144],"showcasing":[145],"improved":[146],"capability":[149],"achieved":[150],"integration":[153],"experts.":[156]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-10T00:00:00"}