{"id":"https://openalex.org/W4404089451","doi":"https://doi.org/10.48550/arxiv.2410.16267","title":"xGen-MM-Vid (BLIP-3-Video): You Only Need 32 Tokens to Represent a Video Even in VLMs","display_name":"xGen-MM-Vid (BLIP-3-Video): You Only Need 32 Tokens to Represent a Video Even in VLMs","publication_year":2024,"publication_date":"2024-10-21","ids":{"openalex":"https://openalex.org/W4404089451","doi":"https://doi.org/10.48550/arxiv.2410.16267"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2410.16267","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.16267","pdf_url":"https://arxiv.org/pdf/2410.16267","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,"listed_in":[],"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":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2410.16267","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5084829008","display_name":"Michael S. Ryoo","orcid":"https://orcid.org/0000-0002-5452-8332","observed_orcids":["https://orcid.org/0000-0002-5452-8332"]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ryoo, Michael S.","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101900985","display_name":"Honglu Zhou","orcid":"https://orcid.org/0000-0002-2953-6277","observed_orcids":["https://orcid.org/0000-0002-2953-6277"]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhou, Honglu","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107421456","display_name":"Shrikant Kendre","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kendre, Shrikant","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113407743","display_name":"Can Qin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qin, Can","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100733668","display_name":"Le Xue","orcid":"https://orcid.org/0000-0003-2810-770X","observed_orcids":["https://orcid.org/0000-0003-2810-770X"]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xue, Le","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015044935","display_name":"Manli Shu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shu, Manli","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Park, Jongwoo","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I59553526","display_name":"Stony Brook University","ror":"https://ror.org/05qghxh33","country_code":"US","type":"education","lineage":["https://openalex.org/I59553526"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Park, Jongwoo","raw_affiliation_strings":["Stony Brook University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stony Brook University","institution_ids":["https://openalex.org/I59553526"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Ranasinghe, Kanchana","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I59553526","display_name":"Stony Brook University","ror":"https://ror.org/05qghxh33","country_code":"US","type":"education","lineage":["https://openalex.org/I59553526"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ranasinghe, Kanchana","raw_affiliation_strings":["Stony Brook University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stony Brook University","institution_ids":["https://openalex.org/I59553526"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042646536","display_name":"Silvio Savarese","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Savarese, Silvio","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100687200","display_name":"Ran Xu","orcid":"https://orcid.org/0000-0003-1387-6696","observed_orcids":["https://orcid.org/0000-0003-1387-6696"]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xu, Ran","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032046813","display_name":"Caiming Xiong","orcid":"https://orcid.org/0000-0003-0349-8628","observed_orcids":["https://orcid.org/0000-0003-0349-8628"]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiong, Caiming","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5018518655","display_name":"Juan Carlos Niebles","orcid":"https://orcid.org/0000-0001-8225-9793","observed_orcids":["https://orcid.org/0000-0001-8225-9793","https://orcid.org/0000-0002-3953-6297"]},"institutions":[{"id":"https://openalex.org/I4210155268","display_name":"Salesforce (United States)","ror":"https://ror.org/057315g56","country_code":"US","type":"company","lineage":["https://openalex.org/I4210155268"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Niebles, Juan Carlos","raw_affiliation_strings":["Salesforce AI Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Salesforce AI Research","institution_ids":["https://openalex.org/I4210155268"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":26,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":1.0,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"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/video-language-models","display_name":"video-language models","score":0.6940000057220459},{"id":"https://openalex.org/keywords/visual-tokenization","display_name":"visual tokenization","score":0.6570000052452087},{"id":"https://openalex.org/keywords/temporal-coding","display_name":"temporal coding","score":0.7369999885559082},{"id":"https://openalex.org/keywords/video-question-answering","display_name":"video question answering","score":0.9139999747276306},{"id":"https://openalex.org/keywords/visual-tokens","display_name":"visual tokens","score":0.3269999921321869}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5376667976379395},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.4702075123786926},{"id":"https://openalex.org/C2988634675","wikidata":"https://www.wikidata.org/wiki/Q34508","display_name":"Video recording","level":2,"score":0.4559401571750641},{"id":"https://openalex.org/C2988454689","wikidata":"https://www.wikidata.org/wiki/Q173131","display_name":"Digital video","level":3,"score":0.43000638484954834},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.34139785170555115},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.10796856880187988},{"id":"https://openalex.org/C761482","wikidata":"https://www.wikidata.org/wiki/Q118093","display_name":"Transmission (telecommunications)","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2410.16267","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.16267","pdf_url":"https://arxiv.org/pdf/2410.16267","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,"listed_in":[],"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":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2410.16267","doi":"https://doi.org/10.48550/arxiv.2410.16267","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2410.16267","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,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2410.16267","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.16267","pdf_url":"https://arxiv.org/pdf/2410.16267","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,"listed_in":[],"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":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4404089451.pdf","grobid_xml":"https://content.openalex.org/works/W4404089451.grobid-xml"},"referenced_works_count":29,"referenced_works":["https://openalex.org/W4229042118","https://openalex.org/W4306886919","https://openalex.org/W4310629611","https://openalex.org/W4379918953","https://openalex.org/W4387225871","https://openalex.org/W4389326242","https://openalex.org/W4389519526","https://openalex.org/W4393212018","https://openalex.org/W4394007474","https://openalex.org/W4400341198","https://openalex.org/W4401306886","https://openalex.org/W4402703549","https://openalex.org/W4402857639","https://openalex.org/W4402954393","https://openalex.org/W4403370767","https://openalex.org/W4403760933","https://openalex.org/W4403939369","https://openalex.org/W4404304985","https://openalex.org/W4404782265","https://openalex.org/W4404784276","https://openalex.org/W4405429615","https://openalex.org/W4406073438","https://openalex.org/W4412888622","https://openalex.org/W4413145669","https://openalex.org/W4413147844","https://openalex.org/W4415798090","https://openalex.org/W7103755807","https://openalex.org/W7108068597","https://openalex.org/W7128429593"],"related_works":["https://openalex.org/W7160833777","https://openalex.org/W6979370130","https://openalex.org/W4403002096","https://openalex.org/W4403574051","https://openalex.org/W7214988040","https://openalex.org/W7092298777","https://openalex.org/W7166669160","https://openalex.org/W7128501944","https://openalex.org/W7204765602","https://openalex.org/W4416036624"],"abstract_inverted_index":{"We":[0,68,88],"present":[1],"xGen-MM-Vid":[2],"(BLIP-3-Video):":[3],"a":[4,36,44],"multimodal":[5],"language":[6],"model":[7],"for":[8],"videos,":[9],"particularly":[10],"designed":[11],"to":[12,29,53,98],"efficiently":[13],"capture":[14],"temporal":[15,73],"information":[16],"over":[17,40],"multiple":[18,41],"frames.":[19],"BLIP-3-Video":[20,92],"takes":[21],"advantage":[22],"of":[23,38,47,72],"the":[24,30],"'temporal":[25],"encoder'":[26],"in":[27],"addition":[28],"conventional":[31],"visual":[32,48,57,117],"tokenizer,":[33],"which":[34],"maps":[35],"sequence":[37],"tokens":[39,58],"frames":[42],"into":[43],"compact":[45],"set":[46],"tokens.":[49,118],"This":[50],"enables":[51],"BLIP3-Video":[52],"use":[54],"much":[55,99,107],"fewer":[56,116],"than":[59],"its":[60],"competing":[61],"models":[62,83,102],"(e.g.,":[63,103],"32":[64],"vs.":[65],"4608":[66],"tokens).":[67],"explore":[69],"different":[70],"types":[71],"encoders,":[74],"including":[75],"learnable":[76],"spatio-temporal":[77],"pooling":[78],"as":[79,81],"well":[80],"sequential":[82],"like":[84],"Token":[85],"Turing":[86],"Machines.":[87],"experimentally":[89],"confirm":[90],"that":[91],"obtains":[93],"video":[94],"question-answering":[95],"accuracies":[96],"comparable":[97],"larger":[100],"state-of-the-art":[101],"34B),":[104],"while":[105],"being":[106],"smaller":[108],"(i.e.,":[109],"4B)":[110],"and":[111],"more":[112],"efficient":[113],"by":[114],"using":[115],"The":[119],"project":[120],"website":[121],"is":[122],"at":[123],"https://www.salesforceairesearch.com/opensource/xGen-MM-Vid/index.html":[124]},"counts_by_year":[{"year":2026,"cited_by_count":13},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":2}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2024-11-06T00:00:00"}