{"id":"https://openalex.org/W4403580262","doi":"https://doi.org/10.1109/iccv51701.2025.01433","title":"PUMA: Empowering Unified MLLM with Multi-Granular Visual Generation","display_name":"PUMA: Empowering Unified MLLM with Multi-Granular Visual Generation","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4403580262","doi":"https://doi.org/10.1109/iccv51701.2025.01433"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.01433","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.01433","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2410.13861","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5025456709","display_name":"Rongyao Fang","orcid":"https://orcid.org/0000-0002-8010-4808"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Rongyao Fang","raw_affiliation_strings":["CUHK MMLab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CUHK MMLab","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054320268","display_name":"Chengqi Duan","orcid":"https://orcid.org/0009-0004-5737-019X"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Chengqi Duan","raw_affiliation_strings":["HKU"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"HKU","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100369590","display_name":"Kun Wang","orcid":"https://orcid.org/0000-0002-0450-3997"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kun Wang","raw_affiliation_strings":["SenseTime"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SenseTime","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100348668","display_name":"Hao Li","orcid":"https://orcid.org/0000-0002-9923-4346"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Hao Li","raw_affiliation_strings":["CUHK MMLab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CUHK MMLab","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071362658","display_name":"Hao Tian","orcid":"https://orcid.org/0000-0002-4810-5798"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Linjiang Huang","raw_affiliation_strings":["BUAA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"BUAA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060379760","display_name":"Xingyu Zeng","orcid":"https://orcid.org/0009-0007-8224-4461"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Tian","raw_affiliation_strings":["SenseTime"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SenseTime","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024522346","display_name":"Rui Zhao","orcid":"https://orcid.org/0000-0001-5874-131X"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xingyu Zeng","raw_affiliation_strings":["SenseTime"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SenseTime","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026944066","display_name":"Jifeng Dai","orcid":"https://orcid.org/0000-0002-6785-0785"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Zhao","raw_affiliation_strings":["SenseTime"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SenseTime","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100732450","display_name":"Hongsheng Li","orcid":"https://orcid.org/0000-0002-2664-7975"},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":null,"type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":[],"is_corresponding":false,"raw_author_name":"Jifeng Dai","raw_affiliation_strings":["Shanghai AI Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai AI Lab","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027234036","display_name":"Xihui Liu","orcid":"https://orcid.org/0000-0002-1791-8745"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Hongsheng Li","raw_affiliation_strings":["CUHK MMLab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CUHK MMLab","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5027234036","display_name":"Xihui Liu","orcid":"https://orcid.org/0000-0002-1791-8745"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xihui Liu","raw_affiliation_strings":["HKU"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"HKU","institution_ids":["https://openalex.org/I889458895"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":11,"corresponding_author_ids":["https://openalex.org/A5025456709"],"corresponding_institution_ids":["https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00056672,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15447","last_page":"15457"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9621999859809875,"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":0.9621999859809875,"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/T11439","display_name":"Video Analysis and Summarization","score":0.9151999950408936,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.9088000059127808,"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/puma","display_name":"Puma","score":0.7500426173210144},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.42758437991142273},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3339548707008362},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.09802520275115967}],"concepts":[{"id":"https://openalex.org/C2777417711","wikidata":"https://www.wikidata.org/wiki/Q270748","display_name":"Puma","level":3,"score":0.7500426173210144},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.42758437991142273},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3339548707008362},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.09802520275115967},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/iccv51701.2025.01433","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.01433","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2410.13861","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.13861","pdf_url":"https://arxiv.org/pdf/2410.13861","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":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:hub.hku.hk:10722/359188","is_oa":false,"landing_page_url":"https://hub.hku.hk/handle/10722/359188","pdf_url":null,"source":{"id":"https://openalex.org/S4377196271","display_name":"The HKU Scholars Hub (University of Hong Kong)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I889458895","host_organization_name":"University of Hong Kong","host_organization_lineage":["https://openalex.org/I889458895"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference_Paper"},{"id":"doi:10.48550/arxiv.2410.13861","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2410.13861","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:2410.13861","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.13861","pdf_url":"https://arxiv.org/pdf/2410.13861","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":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W3029624080","https://openalex.org/W2383247791","https://openalex.org/W2383847661","https://openalex.org/W2977641071","https://openalex.org/W2112968535","https://openalex.org/W2255064539","https://openalex.org/W2347355260"],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,10,55,63,117,153],"multimodal":[3,21,108,122],"foundation":[4],"models":[5,24],"have":[6,15,33],"yielded":[7],"significant":[8,128],"progress":[9],"vision-language":[11],"understanding.":[12],"Initial":[13],"attempts":[14],"also":[16],"explored":[17],"the":[18,36,52,59,93,139],"potential":[19],"of":[20,40,89,97,121,136,142],"large":[22],"language":[23],"(MLLMs)":[25],"for":[26],"visual":[27,77,82,144],"content":[28],"generation.":[29,78],"However,":[30],"existing":[31],"works":[32],"insufficiently":[34],"addressed":[35],"varying":[37],"granularity":[38,95,140],"demands":[39,141],"different":[41,94],"image":[42,64,99],"generation":[43,57,100],"tasks":[44,101],"within":[45,102],"a":[46,103,118,127,131],"unified":[47,104,133],"MLLM":[48,74,105,134],"paradigm":[49],"-":[50],"from":[51],"diversity":[53],"required":[54],"text-to-image":[56],"to":[58,138],"precise":[60],"controllability":[61],"needed":[62],"manipulation.":[65],"In":[66],"this":[67],"work,":[68],"we":[69],"propose":[70],"PUMA,":[71],"emPowering":[72],"Unified":[73],"with":[75],"Multi-grAnular":[76],"PUMA":[79,114],"unifies":[80],"multi-granular":[81],"features":[83],"as":[84],"both":[85],"inputs":[86],"and":[87,110,148],"outputs":[88],"MLLMs,":[90],"elegantly":[91],"addressing":[92],"requirements":[96],"various":[98,143],"framework.":[106],"Following":[107],"pretraining":[109],"task-specific":[111],"instruction":[112],"tuning,":[113],"demonstrates":[115],"proficiency":[116],"wide":[119],"range":[120],"tasks.":[123,145],"This":[124],"work":[125],"represents":[126],"step":[129],"towards":[130],"truly":[132],"capable":[135],"adapting":[137],"The":[146],"code":[147],"model":[149],"will":[150],"be":[151],"released":[152],"https://github.com/rongyaofang/PUMA.":[154]},"counts_by_year":[],"updated_date":"2026-05-06T06:03:25.996018","created_date":"2025-10-10T00:00:00"}