{"id":"https://openalex.org/W4414580774","doi":"https://doi.org/10.48550/arxiv.2505.17726","title":"Slot-MLLM: Object-Centric Visual Tokenization for Multimodal LLM","display_name":"Slot-MLLM: Object-Centric Visual Tokenization for Multimodal LLM","publication_year":2025,"publication_date":"2025-05-23","ids":{"openalex":"https://openalex.org/W4414580774","doi":"https://doi.org/10.48550/arxiv.2505.17726"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2505.17726","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.17726","pdf_url":"https://arxiv.org/pdf/2505.17726","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2505.17726","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5112847265","display_name":"Donghwan Chi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chi, Donghwan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060356922","display_name":"Hyomin Kim","orcid":"https://orcid.org/0000-0002-6350-405X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Hyomin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5105345091","display_name":"Yoonjin Oh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Oh, Yoonjin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100719352","display_name":"Yong-Jin Kim","orcid":"https://orcid.org/0000-0002-9866-2648"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Yongjin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100422970","display_name":"Dong\u2010Hoon Lee","orcid":"https://orcid.org/0000-0002-5013-4440"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Donghoon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033558218","display_name":"Daejin Jo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jo, Daejin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100322020","display_name":"Jongmin Kim","orcid":"https://orcid.org/0000-0003-2937-3073"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Jongmin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Baek, Junyeob","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Baek, Junyeob","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057980326","display_name":"Sungjin Ahn","orcid":"https://orcid.org/0000-0002-3419-8639"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ahn, Sungjin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5087294413","display_name":"Sungwoong Kim","orcid":"https://orcid.org/0000-0002-2676-9454"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Sungwoong","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/T10181","display_name":"Natural Language Processing Techniques","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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","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"}},{"id":"https://openalex.org/T11439","display_name":"Video Analysis and Summarization","score":0.9538999795913696,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9311000108718872,"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/process","display_name":"Process (computing)","score":0.5860000252723694},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5328999757766724},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.47049999237060547},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4551999866962433},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.44609999656677246},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.4453999996185303},{"id":"https://openalex.org/keywords/glyph","display_name":"Glyph (data visualization)","score":0.43070000410079956},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3871999979019165}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8478000164031982},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6086999773979187},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5860000252723694},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5328999757766724},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.47049999237060547},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4551999866962433},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.44609999656677246},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.4453999996185303},{"id":"https://openalex.org/C142816647","wikidata":"https://www.wikidata.org/wiki/Q5573018","display_name":"Glyph (data visualization)","level":3,"score":0.43070000410079956},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4018000066280365},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3871999979019165},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.37720000743865967},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.3650999963283539},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.3472000062465668},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.3073999881744385},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.2809000015258789},{"id":"https://openalex.org/C105842133","wikidata":"https://www.wikidata.org/wiki/Q1899679","display_name":"Visual communication","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.272599995136261},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.2621999979019165}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2505.17726","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.17726","pdf_url":"https://arxiv.org/pdf/2505.17726","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.17726","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.17726","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2505.17726","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.17726","pdf_url":"https://arxiv.org/pdf/2505.17726","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"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":{"Recently,":[0],"multimodal":[1,34],"large":[2],"language":[3],"models":[4],"(MLLMs)":[5],"have":[6,21],"emerged":[7],"as":[8],"a":[9,141],"key":[10],"approach":[11],"in":[12,48],"achieving":[13],"artificial":[14],"general":[15],"intelligence.":[16],"In":[17,103],"particular,":[18,104],"vision-language":[19,163],"MLLMs":[20,58,188],"been":[22],"developed":[23],"to":[24,73,136],"generate":[25,77],"not":[26],"only":[27,61],"text":[28],"but":[29],"also":[30,131],"visual":[31,79,94,124,159],"outputs":[32],"from":[33],"inputs.":[35],"This":[36],"advancement":[37],"requires":[38],"efficient":[39],"image":[40,54,68],"tokens":[41,120],"that":[42,165],"LLMs":[43],"can":[44,121],"process":[45],"effectively":[46,74],"both":[47],"input":[49],"and":[50,112,130,170,189],"output.":[51],"However,":[52],"existing":[53],"tokenization":[55],"methods":[56],"for":[57,101],"typically":[59],"capture":[60],"global":[62],"abstract":[63],"concepts":[64],"or":[65,76],"uniformly":[66],"segmented":[67],"patches,":[69],"restricting":[70],"MLLMs'":[71],"capability":[72],"understand":[75],"detailed":[78,168],"content,":[80],"particularly":[81],"at":[82],"the":[83,107,176,180],"object":[84],"level.":[85],"To":[86],"address":[87],"this":[88,173],"limitation,":[89],"we":[90],"propose":[91],"an":[92],"object-centric":[93,183],"tokenizer":[95],"based":[96,105],"on":[97,106],"Slot":[98],"Attention":[99],"specifically":[100],"MLLMs.":[102],"Q-Former":[108],"encoder,":[109],"diffusion":[110],"decoder,":[111],"residual":[113],"vector":[114],"quantization,":[115],"our":[116],"proposed":[117],"discretized":[118],"slot":[119,184],"encode":[122],"local":[123,167],"details":[125],"while":[126],"maintaining":[127],"high-level":[128],"semantics,":[129],"align":[132],"with":[133,157,187],"textual":[134],"data":[135],"be":[137],"integrated":[138],"seamlessly":[139],"within":[140],"unified":[142],"next-token":[143],"prediction":[144],"framework":[145],"of":[146,179,182],"LLMs.":[147],"The":[148],"resulting":[149],"Slot-MLLM":[150],"demonstrates":[151],"significant":[152],"performance":[153],"improvements":[154],"over":[155],"baselines":[156],"previous":[158],"tokenizers":[160],"across":[161],"various":[162],"tasks":[164],"entail":[166],"comprehension":[169],"generation.":[171],"Notably,":[172],"work":[174],"is":[175],"first":[177],"demonstration":[178],"feasibility":[181],"attention":[185],"performed":[186],"in-the-wild":[190],"natural":[191],"images.":[192]},"counts_by_year":[],"updated_date":"2026-08-18T07:49:30.821534","created_date":"2025-10-10T00:00:00"}