{"id":"https://openalex.org/W7212296012","doi":"https://doi.org/10.48550/arxiv.2609.11058","title":"EMMI: Edge Multi-Modal Intelligence for Communication-Efficient MLLM Inference via Fused Representation Compression","display_name":"EMMI: Edge Multi-Modal Intelligence for Communication-Efficient MLLM Inference via Fused Representation Compression","publication_year":2026,"publication_date":"2026-09-10","ids":{"openalex":"https://openalex.org/W7212296012","doi":"https://doi.org/10.48550/arxiv.2609.11058"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.11058","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.11058","pdf_url":"https://arxiv.org/pdf/2609.11058","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/2609.11058","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5074308655","display_name":"Motahare Mounesan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mounesan, Motahare","raw_affiliation_strings":["Texas A&M University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas A&M University","institution_ids":["https://openalex.org/I91045830"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5151308084","display_name":"Irfan Khan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Khan, Irfan","raw_affiliation_strings":["Texas A&M University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas A&M University","institution_ids":["https://openalex.org/I91045830"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I91045830"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9887999892234802,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9887999892234802,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.005799999926239252,"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":0.0017999999690800905,"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/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9750000238418579},{"id":"https://openalex.org/keywords/edge-inference","display_name":"edge inference","score":0.890999972820282},{"id":"https://openalex.org/keywords/communication-efficiency","display_name":"communication efficiency","score":0.8560000061988831},{"id":"https://openalex.org/keywords/representation-compression","display_name":"representation compression","score":0.9409999847412109},{"id":"https://openalex.org/keywords/data-privacy","display_name":"data privacy","score":0.5789999961853027},{"id":"https://openalex.org/keywords/inference-latency-reduction","display_name":"inference latency reduction","score":0.6779999732971191},{"id":"https://openalex.org/keywords/resource-constrained-edge-devices","display_name":"resource-constrained edge devices","score":0.8190000057220459}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7440999746322632},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.737500011920929},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.6504999995231628},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.602400004863739},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5773000121116638},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.5206000208854675},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4778999984264374},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46230000257492065},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.41670000553131104},{"id":"https://openalex.org/C134066672","wikidata":"https://www.wikidata.org/wiki/Q1424639","display_name":"Payload (computing)","level":3,"score":0.4077000021934509},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.3984000086784363},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.39410001039505005},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3709000051021576},{"id":"https://openalex.org/C157170001","wikidata":"https://www.wikidata.org/wiki/Q4781507","display_name":"Applications of artificial intelligence","level":2,"score":0.33149999380111694},{"id":"https://openalex.org/C116409475","wikidata":"https://www.wikidata.org/wiki/Q1385056","display_name":"External Data Representation","level":2,"score":0.31839999556541443},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3037000000476837},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.29269999265670776},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2874999940395355},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.2816999852657318},{"id":"https://openalex.org/C12269588","wikidata":"https://www.wikidata.org/wiki/Q132364","display_name":"Communications protocol","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C193536780","wikidata":"https://www.wikidata.org/wiki/Q1513153","display_name":"Edge detection","level":4,"score":0.2791000008583069}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.11058","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.11058","pdf_url":"https://arxiv.org/pdf/2609.11058","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.2609.11058","doi":"https://doi.org/10.48550/arxiv.2609.11058","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.11058","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":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.11058","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.11058","pdf_url":"https://arxiv.org/pdf/2609.11058","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/W7212296012.pdf","grobid_xml":"https://content.openalex.org/works/W7212296012.grobid-xml"},"referenced_works_count":7,"referenced_works":["https://openalex.org/W1959608418","https://openalex.org/W4297808394","https://openalex.org/W4390437694","https://openalex.org/W4392120646","https://openalex.org/W4406489046","https://openalex.org/W4415747953","https://openalex.org/W6967017503"],"related_works":["https://openalex.org/W7159766555","https://openalex.org/W7150916684","https://openalex.org/W4416109865","https://openalex.org/W4414745519","https://openalex.org/W7169836658","https://openalex.org/W7166716432","https://openalex.org/W7161173582","https://openalex.org/W7131314377","https://openalex.org/W4417358291","https://openalex.org/W7146946737"],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2,157,163],"multimodal":[3,139],"large":[4],"language":[5],"mod-":[6],"els":[7],"(MLLMs)":[8],"have":[9],"opened":[10],"new":[11],"opportunities":[12],"for":[13,109],"edge":[14,36,74,81,130,170],"intelligence":[15],"by":[16,149],"enabling":[17,79],"reasoning":[18],"across":[19],"heterogeneous":[20,129],"sensor":[21,56],"modalities,":[22],"such":[23],"as":[24],"vision,":[25],"text,":[26],"and":[27,46,76,94,123,132],"telemetry":[28],"data.":[29],"However,":[30],"deploying":[31],"these":[32],"capabilities":[33],"on":[34,136],"resource-constrained":[35],"platforms":[37],"remains":[38],"challenging":[39],"due":[40],"to":[41,106,159],"the":[42,98,146],"substantial":[43],"computational,":[44],"memory,":[45],"communication":[47,117,147],"demands":[48],"of":[49],"modern":[50],"MLLMs.":[51,134],"Rather":[52],"than":[53],"transmitting":[54,100],"raw":[55],"observations":[57],"or":[58],"partitioning":[59],"neural":[60],"networks":[61],"at":[62,97],"intermediate":[63],"layers,":[64],"Edge":[65],"Multi-Modal":[66],"Intelligence":[67],"(EMMI)":[68],"communicates":[69],"a":[70,102,125,137,160],"compact":[71,103],"representation":[72,92,105],"between":[73,128],"devices":[75,131],"server":[77],"resources,":[78],"communication-efficient":[80],"MLLM":[82,111],"inference.":[83],"To":[84],"achieve":[85],"this,":[86],"EMMI":[87,143],"performs":[88],"modality-specific":[89],"encoding,":[90],"cross-modal":[91],"fusion,":[93],"learned":[95],"compression":[96],"edge,":[99],"only":[101],"latent":[104],"server-side":[107,133],"resources":[108],"high-capacity":[110],"reasoning.":[112],"This":[113],"representation-centric":[114],"design":[115],"reduces":[116],"overhead,":[118],"preserves":[119],"local":[120],"data":[121],"privacy,":[122],"provides":[124],"fixed-size":[126],"interface":[127],"Evaluation":[135],"representative":[138],"benchmark":[140],"demonstrates":[141],"that":[142],"can":[144],"reduce":[145],"payload":[148],"32x":[150],"while":[151],"maintaining":[152],"comparable":[153],"downstream":[154],"accuracy,":[155],"resulting":[156],"up":[158],"3.4x":[161],"reduction":[162],"estimated":[164],"end-to-end":[165],"inference":[166],"latency":[167],"under":[168],"bandwidth-constrained":[169],"conditions.":[171]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-12T00:00:00"}