{"id":"https://openalex.org/W4413120574","doi":"https://doi.org/10.1109/mm.2025.3596539","title":"Characterizing and Efficiently Accelerating Multimodal Generation Model Inference","display_name":"Characterizing and Efficiently Accelerating Multimodal Generation Model Inference","publication_year":2025,"publication_date":"2025-08-08","ids":{"openalex":"https://openalex.org/W4413120574","doi":"https://doi.org/10.1109/mm.2025.3596539"},"language":"en","primary_location":{"id":"doi:10.1109/mm.2025.3596539","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mm.2025.3596539","pdf_url":null,"source":{"id":"https://openalex.org/S59697426","display_name":"IEEE Micro","issn_l":"0272-1732","issn":["0272-1732","1937-4143"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Micro","raw_type":"journal-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":null,"display_name":"Yejin Lee","orcid":"https://orcid.org/0000-0001-9532-1621"},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yejin Lee","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","AI Research at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"AI Research at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104179257","display_name":"Alicia Golden","orcid":null},"institutions":[{"id":"https://openalex.org/I4210100866","display_name":"Visiting Nurse Association","ror":"https://ror.org/0127psw24","country_code":"US","type":"other","lineage":["https://openalex.org/I4210100866"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alicia Golden","raw_affiliation_strings":["Meta, Cambridge, MA, USA","Visiting Researcher at Meta, Cambridge, MA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Cambridge, MA, USA","institution_ids":[]},{"raw_affiliation_string":"Visiting Researcher at Meta, Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210100866"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085022206","display_name":"Anna Sun","orcid":"https://orcid.org/0000-0002-6212-1313"},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210165293","display_name":"Metacomp Technologies (United States)","ror":"https://ror.org/05cxavv95","country_code":"US","type":"company","lineage":["https://openalex.org/I4210165293"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anna Sun","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Engineer at Meta, San Francisco, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Engineer at Meta, San Francisco, CA, USA","institution_ids":["https://openalex.org/I4210165293"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077177791","display_name":"Basil Hosmer","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Basil Hosmer","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Scientist at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Scientist at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005070359","display_name":"Bilge Acun","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bilge Acun","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Scientist at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Scientist at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042895564","display_name":"Can Balioglu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210112286","display_name":"The Metropolitan Opera (United States)","ror":"https://ror.org/022q3vw94","country_code":"US","type":"company","lineage":["https://openalex.org/I4210112286"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Can Balioglu","raw_affiliation_strings":["Meta, New York, NY, USA","Research Engineer at Meta, New York, USA"],"affiliations":[{"raw_affiliation_string":"Meta, New York, NY, USA","institution_ids":[]},{"raw_affiliation_string":"Research Engineer at Meta, New York, USA","institution_ids":["https://openalex.org/I4210112286"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087491225","display_name":"Changhan Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210112286","display_name":"The Metropolitan Opera (United States)","ror":"https://ror.org/022q3vw94","country_code":"US","type":"company","lineage":["https://openalex.org/I4210112286"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Changhan Wang","raw_affiliation_strings":["Meta, New York, NY, USA","Research Engineer at Meta, New York, USA"],"affiliations":[{"raw_affiliation_string":"Meta, New York, NY, USA","institution_ids":[]},{"raw_affiliation_string":"Research Engineer at Meta, New York, USA","institution_ids":["https://openalex.org/I4210112286"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Charles David Hernandez","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Charles David Hernandez","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Engineer at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Engineer at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029859679","display_name":"Christian Puhrsch","orcid":"https://orcid.org/0009-0002-3925-967X"},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210136555","display_name":"ATUM (United States)","ror":"https://ror.org/03vcdbg75","country_code":"US","type":"company","lineage":["https://openalex.org/I4210136555"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Christian Puhrsch","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Machine Learning Engineer at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Machine Learning Engineer at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210136555"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063864170","display_name":"Daniel Haziza","orcid":null},"institutions":[{"id":"https://openalex.org/I117807411","display_name":"National Society of Professional Engineers","ror":"https://ror.org/01hpjfm62","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I117807411"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Daniel Haziza","raw_affiliation_strings":["Meta, Paris, France","Research Engineer at Meta, Paris, France"],"affiliations":[{"raw_affiliation_string":"Meta, Paris, France","institution_ids":[]},{"raw_affiliation_string":"Research Engineer at Meta, Paris, France","institution_ids":["https://openalex.org/I117807411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114426750","display_name":"Driss Guessous","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210136555","display_name":"ATUM (United States)","ror":"https://ror.org/03vcdbg75","country_code":"US","type":"company","lineage":["https://openalex.org/I4210136555"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Driss Guessous","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Machine Learning Engineer at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Machine Learning Engineer at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210136555"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032569443","display_name":"Francisco Massa","orcid":"https://orcid.org/0000-0003-0697-6664"},"institutions":[{"id":"https://openalex.org/I117807411","display_name":"National Society of Professional Engineers","ror":"https://ror.org/01hpjfm62","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I117807411"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Francisco Massa","raw_affiliation_strings":["Meta, Paris, France","Research Engineer at Meta, Paris, France"],"affiliations":[{"raw_affiliation_string":"Meta, Paris, France","institution_ids":[]},{"raw_affiliation_string":"Research Engineer at Meta, Paris, France","institution_ids":["https://openalex.org/I117807411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112136908","display_name":"Jacob Kahn","orcid":null},"institutions":[{"id":"https://openalex.org/I117807411","display_name":"National Society of Professional Engineers","ror":"https://ror.org/01hpjfm62","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I117807411"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jacob Kahn","raw_affiliation_strings":["Meta, New York, NY, USA","Research Engineer manager at Meta, New York, USA"],"affiliations":[{"raw_affiliation_string":"Meta, New York, NY, USA","institution_ids":[]},{"raw_affiliation_string":"Research Engineer manager at Meta, New York, USA","institution_ids":["https://openalex.org/I117807411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110584158","display_name":"Jeffrey K. S. Wan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210160740","display_name":"Integrated Software (United States)","ror":"https://ror.org/05dp7m259","country_code":"US","type":"company","lineage":["https://openalex.org/I4210160740"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jeffrey Wan","raw_affiliation_strings":["Meta, New York, NY, USA","Software Engineer at Meta, New York, USA"],"affiliations":[{"raw_affiliation_string":"Meta, New York, NY, USA","institution_ids":[]},{"raw_affiliation_string":"Software Engineer at Meta, New York, USA","institution_ids":["https://openalex.org/I4210160740"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082867662","display_name":"Jeremy Reizenstein","orcid":null},"institutions":[{"id":"https://openalex.org/I2800804717","display_name":"Institution of Civil Engineers","ror":"https://ror.org/0402pb169","country_code":"GB","type":"nonprofit","lineage":["https://openalex.org/I2800804717"]},{"id":"https://openalex.org/I4210111288","display_name":"Meta (United Kingdom)","ror":"https://ror.org/020ye1821","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210111288","https://openalex.org/I4210114444"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Jeremy Reizenstein","raw_affiliation_strings":["Meta, London, U.K","Research Engineer at Meta, London, UK"],"affiliations":[{"raw_affiliation_string":"Meta, London, U.K","institution_ids":["https://openalex.org/I4210111288"]},{"raw_affiliation_string":"Research Engineer at Meta, London, UK","institution_ids":["https://openalex.org/I4210111288","https://openalex.org/I2800804717"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100349821","display_name":"Jiaqi Zhai","orcid":"https://orcid.org/0009-0004-7279-3318"},"institutions":[{"id":"https://openalex.org/I174135032","display_name":"Bellevue College","ror":"https://ror.org/05gr4yv49","country_code":"US","type":"education","lineage":["https://openalex.org/I174135032"]},{"id":"https://openalex.org/I4210160740","display_name":"Integrated Software (United States)","ror":"https://ror.org/05dp7m259","country_code":"US","type":"company","lineage":["https://openalex.org/I4210160740"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiaqi Zhai","raw_affiliation_strings":["Meta, Bellevue, WA, USA","Software Engineer at Meta, New York, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Bellevue, WA, USA","institution_ids":["https://openalex.org/I174135032"]},{"raw_affiliation_string":"Software Engineer at Meta, New York, USA","institution_ids":["https://openalex.org/I4210160740"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Joe Isaacson","orcid":"https://orcid.org/0000-0002-8220-4693"},"institutions":[{"id":"https://openalex.org/I58610484","display_name":"Seattle University","ror":"https://ror.org/02jqc0m91","country_code":"US","type":"education","lineage":["https://openalex.org/I58610484"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Joe Isaacson","raw_affiliation_strings":["Meta, Seattle, WA, USA","Engineering Leader at Meta, Seattle, WA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Seattle, WA, USA","institution_ids":[]},{"raw_affiliation_string":"Engineering Leader at Meta, Seattle, WA, USA","institution_ids":["https://openalex.org/I58610484"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040869952","display_name":"Joel Schlosser","orcid":null},"institutions":[{"id":"https://openalex.org/I4210136555","display_name":"ATUM (United States)","ror":"https://ror.org/03vcdbg75","country_code":"US","type":"company","lineage":["https://openalex.org/I4210136555"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Joel Schlosser","raw_affiliation_strings":["Meta, New York, NY, USA","Machine Learning Engineer at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, New York, NY, USA","institution_ids":[]},{"raw_affiliation_string":"Machine Learning Engineer at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210136555"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058915697","display_name":"Juan Pino","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]},{"id":"https://openalex.org/I4210165293","display_name":"Metacomp Technologies (United States)","ror":"https://ror.org/05cxavv95","country_code":"US","type":"company","lineage":["https://openalex.org/I4210165293"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Juan Pino","raw_affiliation_strings":["Meta, San Francisco, CA, USA","Researcher at Meta, Menlo Park, USA"],"affiliations":[{"raw_affiliation_string":"Meta, San Francisco, CA, USA","institution_ids":["https://openalex.org/I4210165293"]},{"raw_affiliation_string":"Researcher at Meta, Menlo Park, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023501975","display_name":"Kaushik Ram Sadagopan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kaushik Ram Sadagopan","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Engineer at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Engineer at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107673193","display_name":"Leonid Shamis","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Leonid Shamis","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Software Engineer at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Software Engineer at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084195463","display_name":"Linjian Ma","orcid":"https://orcid.org/0000-0001-7470-5415"},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Linjian Ma","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Scientist at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Scientist at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055817509","display_name":"Min-Jae Hwang","orcid":"https://orcid.org/0000-0002-7376-009X"},"institutions":[{"id":"https://openalex.org/I4210100094","display_name":"Metro Transit","ror":"https://ror.org/012wk3x67","country_code":"US","type":"government","lineage":["https://openalex.org/I4210100094","https://openalex.org/I4405261780"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Min-Jae Hwang","raw_affiliation_strings":["Meta, Seattle, WA, USA","Research Scientist at Meta, Seattle, WA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Seattle, WA, USA","institution_ids":[]},{"raw_affiliation_string":"Research Scientist at Meta, Seattle, WA, USA","institution_ids":["https://openalex.org/I4210100094"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070933659","display_name":"Mingda Chen","orcid":"https://orcid.org/0000-0002-1824-5263"},"institutions":[{"id":"https://openalex.org/I4210112286","display_name":"The Metropolitan Opera (United States)","ror":"https://ror.org/022q3vw94","country_code":"US","type":"company","lineage":["https://openalex.org/I4210112286"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mingda Chen","raw_affiliation_strings":["Meta, New York, NY, USA","Research Scientist at Meta, NYC, NY, USA"],"affiliations":[{"raw_affiliation_string":"Meta, New York, NY, USA","institution_ids":[]},{"raw_affiliation_string":"Research Scientist at Meta, NYC, NY, USA","institution_ids":["https://openalex.org/I4210112286"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031013766","display_name":"Mostafa Elhoushi","orcid":"https://orcid.org/0000-0001-6172-4510"},"institutions":[{"id":"https://openalex.org/I4210112286","display_name":"The Metropolitan Opera (United States)","ror":"https://ror.org/022q3vw94","country_code":"US","type":"company","lineage":["https://openalex.org/I4210112286"]},{"id":"https://openalex.org/I4210140397","display_name":"Metrica (United States)","ror":"https://ror.org/031ffw737","country_code":"US","type":"company","lineage":["https://openalex.org/I4210140397"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mostafa Elhoushi","raw_affiliation_strings":["Meta, Toronto, ON, Canada","Research Engineer at Meta, New York, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Toronto, ON, Canada","institution_ids":["https://openalex.org/I4210140397"]},{"raw_affiliation_string":"Research Engineer at Meta, New York, USA","institution_ids":["https://openalex.org/I4210112286"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100707159","display_name":"Pedro Rodr\u00edguez","orcid":"https://orcid.org/0000-0001-8572-0725"},"institutions":[{"id":"https://openalex.org/I4210100094","display_name":"Metro Transit","ror":"https://ror.org/012wk3x67","country_code":"US","type":"government","lineage":["https://openalex.org/I4210100094","https://openalex.org/I4405261780"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pedro Rodriguez","raw_affiliation_strings":["Meta, Seattle, WA, USA","Research Scientist at Meta, Seattle, WA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Seattle, WA, USA","institution_ids":[]},{"raw_affiliation_string":"Research Scientist at Meta, Seattle, WA, USA","institution_ids":["https://openalex.org/I4210100094"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072261873","display_name":"Ram Pasunuru","orcid":null},"institutions":[{"id":"https://openalex.org/I4210100094","display_name":"Metro Transit","ror":"https://ror.org/012wk3x67","country_code":"US","type":"government","lineage":["https://openalex.org/I4210100094","https://openalex.org/I4405261780"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ram Pasunuru","raw_affiliation_strings":["Meta, Seattle, WA, USA","Research Scientist at Meta, Seattle, WA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Seattle, WA, USA","institution_ids":[]},{"raw_affiliation_string":"Research Scientist at Meta, Seattle, WA, USA","institution_ids":["https://openalex.org/I4210100094"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047381366","display_name":"Samuel Hsia","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Samuel Hsia","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Scientist at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Scientist at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027596907","display_name":"Scott Yih","orcid":null},"institutions":[{"id":"https://openalex.org/I4210100094","display_name":"Metro Transit","ror":"https://ror.org/012wk3x67","country_code":"US","type":"government","lineage":["https://openalex.org/I4210100094","https://openalex.org/I4405261780"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Scott Yih","raw_affiliation_strings":["Meta, Seattle, WA, USA","Research Scientist at Meta, Seattle, WA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Seattle, WA, USA","institution_ids":[]},{"raw_affiliation_string":"Research Scientist at Meta, Seattle, WA, USA","institution_ids":["https://openalex.org/I4210100094"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059951425","display_name":"Sravya Popuri","orcid":null},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sravya Popuri","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Manager at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Manager at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xing Liu","orcid":"https://orcid.org/0009-0005-0172-8698"},"institutions":[{"id":"https://openalex.org/I4210099336","display_name":"Menlo School","ror":"https://ror.org/01240pn49","country_code":"US","type":"education","lineage":["https://openalex.org/I4210099336"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xing Liu","raw_affiliation_strings":["Meta, Menlo Park, CA, USA","Research Scientist at Meta, Menlo Park, CA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210099336"]},{"raw_affiliation_string":"Research Scientist at Meta, Menlo Park, CA, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5028220093","display_name":"Carole-Jean Wu","orcid":"https://orcid.org/0000-0002-9032-7239"},"institutions":[{"id":"https://openalex.org/I2801348669","display_name":"Medieval Academy of America","ror":"https://ror.org/04g3e3p90","country_code":"US","type":"other","lineage":["https://openalex.org/I2801348669"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Carole-Jean Wu","raw_affiliation_strings":["Meta, Cambridge, MA, USA","Director of AI Research at Meta, Cambridge, MA, USA"],"affiliations":[{"raw_affiliation_string":"Meta, Cambridge, MA, USA","institution_ids":[]},{"raw_affiliation_string":"Director of AI Research at Meta, Cambridge, MA, USA","institution_ids":["https://openalex.org/I2801348669"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":32,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210099336","https://openalex.org/I4210114444"],"apc_list":null,"apc_paid":null,"fwci":2.26,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.90204042,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":"45","issue":"5","first_page":"103","last_page":"115"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.984499990940094,"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/T12031","display_name":"Speech and dialogue systems","score":0.984499990940094,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9532999992370605,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9430000185966492,"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":0.8762450218200684},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.622630774974823},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3826275169849396},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.35120633244514465}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8762450218200684},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.622630774974823},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3826275169849396},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.35120633244514465}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/mm.2025.3596539","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mm.2025.3596539","pdf_url":null,"source":{"id":"https://openalex.org/S59697426","display_name":"IEEE Micro","issn_l":"0272-1732","issn":["0272-1732","1937-4143"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Micro","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W4394998532","https://openalex.org/W4399377323","https://openalex.org/W4402671987","https://openalex.org/W4413120574"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Generative":[0],"artificial":[1],"intelligence":[2],"(AI)":[3],"technology":[4],"is":[5,77],"revolutionizing":[6],"the":[7,45,95,107],"computing":[8],"industry,":[9],"posing":[10],"new":[11],"system":[12,31,57,124],"design":[13,58],"and":[14,23,51,59,126],"optimization":[15,60,118],"opportunities.":[16],"In":[17,89],"particular,":[18],"AI\u2019s":[19],"ability":[20],"to":[21,40,91,106,123],"understand":[22],"respond":[24],"in":[25,44,111],"multiple":[26],"modalities":[27],"comes":[28],"with":[29],"significant":[30,102],"resource":[32],"demands.":[33],"To":[34],"sustainably":[35],"scale":[36],"generative":[37,96],"AI":[38,97],"capabilities":[39],"billions":[41],"of":[42,66],"users":[43],"world,":[46],"inference":[47,103],"must":[48],"be":[49],"fast":[50],"efficient.":[52],"This":[53],"article":[54],"pinpoints":[55],"key":[56],"opportunities":[61],"by":[62,85],"characterizing":[63],"a":[64,78,129],"family":[65],"emerging":[67],"multimodal":[68],"generation":[69,76],"models":[70],"on":[71],"real":[72],"systems.":[73],"Autoregressive":[74],"token":[75],"critical":[79],"latency":[80,104],"performance":[81],"bottleneck,":[82],"typically":[83],"dominated":[84],"GPU":[86],"idle":[87],"time.":[88],"addition":[90],"memory-intensive":[92],"attention":[93],"across":[94],"models,":[98],"linear":[99],"operations":[100],"constitute":[101],"due":[105],"feed":[108],"forward":[109],"networks":[110],"Transformer-based":[112],"models.":[113],"We":[114],"demonstrate":[115],"that":[116],"state-of-the-art":[117],"levers,":[119],"spanning":[120],"from":[121],"applications":[122],"software":[125],"hardware,":[127],"set":[128],"3.88\u00d7":[130],"better":[131],"baseline.":[132]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-04-17T18:11:37.981687","created_date":"2025-10-10T00:00:00"}
