{"id":"https://openalex.org/W4414197493","doi":"https://doi.org/10.1109/dac63849.2025.11132644","title":"EdgeMM: Multi-Core CPU with Heterogeneous AI-Extension and Activation-aware Weight Pruning for Multimodal LLMs at Edge","display_name":"EdgeMM: Multi-Core CPU with Heterogeneous AI-Extension and Activation-aware Weight Pruning for Multimodal LLMs at Edge","publication_year":2025,"publication_date":"2025-06-22","ids":{"openalex":"https://openalex.org/W4414197493","doi":"https://doi.org/10.1109/dac63849.2025.11132644"},"language":"en","primary_location":{"id":"doi:10.1109/dac63849.2025.11132644","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132644","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","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/A5111301706","display_name":"Kangbo Bai","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Kangbo Bai","raw_affiliation_strings":["School of Integrated Circuits Peking University,Beijing,China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits Peking University,Beijing,China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088551028","display_name":"Tianyu Jia","orcid":"https://orcid.org/0000-0002-4570-4613"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Le Ye","raw_affiliation_strings":["School of Integrated Circuits Peking University,Beijing,China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits Peking University,Beijing,China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012603707","display_name":"Ru Huang","orcid":"https://orcid.org/0000-0001-7545-0987"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ru Huang","raw_affiliation_strings":["School of Integrated Circuits Peking University,Beijing,China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits Peking University,Beijing,China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003039083","display_name":"Le Ye","orcid":"https://orcid.org/0000-0003-0599-7762"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tianyu Jia","raw_affiliation_strings":["School of Integrated Circuits Peking University,Beijing,China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits Peking University,Beijing,China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5111301706"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13440473,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.8307999968528748,"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":0.8307999968528748,"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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.7462000250816345,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T12535","display_name":"Machine Learning and Data Classification","score":0.7398999929428101,"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/laptop","display_name":"Laptop","score":0.6363999843597412},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5735999941825867},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5181000232696533},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.48190000653266907},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4138999879360199},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.4065999984741211},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.40220001339912415},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.39739999175071716}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6988999843597412},{"id":"https://openalex.org/C2780008327","wikidata":"https://www.wikidata.org/wiki/Q3962","display_name":"Laptop","level":2,"score":0.6363999843597412},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5735999941825867},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5181000232696533},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.48190000653266907},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4138999879360199},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4097999930381775},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.4065999984741211},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.40220001339912415},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.39739999175071716},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.39419999718666077},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.3935999870300293},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.38280001282691956},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.32249999046325684},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.31119999289512634},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.30630001425743103},{"id":"https://openalex.org/C2781137444","wikidata":"https://www.wikidata.org/wiki/Q237105","display_name":"Sharpening","level":2,"score":0.299699991941452},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2759000062942505},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26899999380111694},{"id":"https://openalex.org/C2776096036","wikidata":"https://www.wikidata.org/wiki/Q1140483","display_name":"Narrowband","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.25760000944137573},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2558000087738037},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac63849.2025.11132644","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132644","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W3013692244","https://openalex.org/W3016211260","https://openalex.org/W4220980777","https://openalex.org/W4286571858","https://openalex.org/W4360605545","https://openalex.org/W4386763332","https://openalex.org/W4390873312","https://openalex.org/W4394862623","https://openalex.org/W4401211635","https://openalex.org/W4402727495","https://openalex.org/W4402727536","https://openalex.org/W4402727764","https://openalex.org/W4403081466","https://openalex.org/W4403940533","https://openalex.org/W4409568286","https://openalex.org/W4412722243"],"related_works":[],"abstract_inverted_index":{"Emerging":[0],"multimodal":[1],"LLMs":[2],"(MLLMs)":[3],"exhibit":[4],"strong":[5],"cross-modality":[6],"perception":[7],"and":[8,11,29,75,84],"reasoning":[9],"capabilities":[10],"hold":[12],"great":[13],"potential":[14],"for":[15,38],"various":[16],"applications":[17],"at":[18],"edge.":[19],"However,":[20],"MLLMs":[21],"typically":[22],"consist":[23],"of":[24],"a":[25,30,46],"compute-intensive":[26],"modality":[27],"encoder":[28],"memory-bound":[31],"LLM":[32],"decoder,":[33],"leading":[34],"to":[35,80,112],"distinct":[36],"bottlenecks":[37],"hardware":[39],"designs.":[40],"In":[41,69],"this":[42],"work,":[43],"we":[44],"present":[45],"multi-core":[47],"CPU":[48],"solution":[49,93],"with":[50],"heterogeneous":[51],"AI":[52],"extensions,":[53],"which":[54],"are":[55,78],"based":[56],"on":[57],"either":[58],"the":[59],"compute-centric":[60],"systolic":[61],"array":[62],"or":[63],"memory-centric":[64],"digital":[65],"compute-in-memory":[66],"(CIM)":[67],"coprocessors.":[68],"addition,":[70],"dynamic":[71],"activation-aware":[72],"weight":[73],"pruning":[74],"bandwidth":[76,82],"management":[77],"developed":[79],"enhance":[81],"efficiency":[83],"core":[85],"utilization,":[86],"improving":[87],"overall":[88],"performance.":[89],"We":[90],"implemented":[91],"our":[92,101],"using":[94],"commercial":[95],"22nm":[96],"technology.":[97],"For":[98],"representative":[99],"MLLMs,":[100],"evaluations":[102],"show":[103],"EdgeMM":[104],"can":[105],"achieve":[106],"$2.84":[107],"\\times$":[108],"performance":[109],"speedup":[110],"compared":[111],"laptop":[113],"3060":[114],"GPU.":[115]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-10T00:00:00"}