{"id":"https://openalex.org/W7214925253","doi":"https://doi.org/10.48550/arxiv.2609.37090","title":"Task-Oriented Visual Feature Compression via Residual Vector Quantization for Device-Edge Multimodal Inference","display_name":"Task-Oriented Visual Feature Compression via Residual Vector Quantization for Device-Edge Multimodal Inference","publication_year":2026,"publication_date":"2026-09-29","ids":{"openalex":"https://openalex.org/W7214925253","doi":"https://doi.org/10.48550/arxiv.2609.37090"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.37090","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37090","pdf_url":"https://arxiv.org/pdf/2609.37090","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/2609.37090","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5001565375","display_name":"Luning Pang","orcid":"https://orcid.org/0000-0001-5879-6804","observed_orcids":["https://orcid.org/0000-0001-5879-6804"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Pang, Luning","raw_affiliation_strings":["Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153580935","display_name":"Cheng Yuan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuan, Cheng","raw_affiliation_strings":["Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102806605","display_name":"Jiawei Shao","orcid":"https://orcid.org/0000-0001-8836-1430","observed_orcids":["https://orcid.org/0000-0001-8836-1430"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shao, Jiawei","raw_affiliation_strings":["Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014691822","display_name":"Mingtao Huang","orcid":"https://orcid.org/0000-0002-8367-569X","observed_orcids":["https://orcid.org/0000-0002-8367-569X"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huang, Mingtao","raw_affiliation_strings":["Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5052022779","display_name":"Yuan Shen","orcid":"https://orcid.org/0000-0002-9396-1964","observed_orcids":["https://orcid.org/0000-0002-9396-1964"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shen, Yuan","raw_affiliation_strings":["Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering , and Beijing National Research Center for In- formation Science and Technology , Tsinghua University , Beijing 100084 , China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I99065089"],"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":0.46309998631477356,"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":0.46309998631477356,"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":0.3043999969959259,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.10490000247955322,"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/residual-vector-quantization","display_name":"residual vector quantization","score":0.9679999947547913},{"id":"https://openalex.org/keywords/large-multimodal-models","display_name":"large multimodal models","score":0.8119999766349792},{"id":"https://openalex.org/keywords/quantization-error-compensation","display_name":"quantization error compensation","score":0.8169999718666077},{"id":"https://openalex.org/keywords/low-latency-inference","display_name":"low-latency inference","score":0.5400000214576721},{"id":"https://openalex.org/keywords/resource-constrained-devices","display_name":"resource-constrained devices","score":0.6610000133514404}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7720999717712402},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6193000078201294},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.5533000230789185},{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.5414999723434448},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.49000000953674316},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4860999882221222},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.47839999198913574},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.47749999165534973},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4505000114440918},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.44339999556541443},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.397599995136261},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.39730000495910645},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.39469999074935913},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.37369999289512634},{"id":"https://openalex.org/C64185310","wikidata":"https://www.wikidata.org/wiki/Q843483","display_name":"Rate\u2013distortion theory","level":3,"score":0.3700000047683716},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.35569998621940613},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3346000015735626},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.3212999999523163},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3107999861240387},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.2996000051498413},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.27790001034736633},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.2754000127315521},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.27489998936653137},{"id":"https://openalex.org/C134066672","wikidata":"https://www.wikidata.org/wiki/Q1424639","display_name":"Payload (computing)","level":3,"score":0.27309998869895935},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2596000134944916}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.37090","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37090","pdf_url":"https://arxiv.org/pdf/2609.37090","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.2609.37090","doi":"https://doi.org/10.48550/arxiv.2609.37090","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.37090","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:2609.37090","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37090","pdf_url":"https://arxiv.org/pdf/2609.37090","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/W7214925253.pdf","grobid_xml":"https://content.openalex.org/works/W7214925253.grobid-xml"},"referenced_works_count":7,"referenced_works":["https://openalex.org/W2242818861","https://openalex.org/W4307323391","https://openalex.org/W4403370767","https://openalex.org/W4404570200","https://openalex.org/W4404652925","https://openalex.org/W4407794264","https://openalex.org/W7160866125"],"related_works":["https://openalex.org/W4415747953","https://openalex.org/W7170769160","https://openalex.org/W7215000140","https://openalex.org/W7214987093","https://openalex.org/W4417466235","https://openalex.org/W7154055427","https://openalex.org/W4401990618","https://openalex.org/W7146946737","https://openalex.org/W7091187001","https://openalex.org/W7161292544"],"abstract_inverted_index":{"Large":[0],"multimodal":[1,72,129],"models":[2],"(LMMs)":[3],"support":[4],"diverse":[5],"visual":[6,28,136],"understanding":[7],"and":[8,47,55,96,111],"reasoning":[9],"tasks":[10],"but":[11],"are":[12],"often":[13],"impractical":[14],"to":[15,80,100,118,141],"run":[16],"entirely":[17],"on":[18,127],"resource-constrained":[19],"devices.":[20],"Device-edge":[21],"co-inference":[22],"reduces":[23,41,134],"device":[24],"computation,":[25],"yet":[26],"transmitting":[27],"data":[29],"over":[30],"bandwidth-limited":[31],"uplinks":[32],"can":[33],"introduce":[34],"substantial":[35],"delay.":[36],"Task-oriented":[37],"feature":[38,45,67,84,109],"compression":[39,68],"(TOFC)":[40],"the":[42,120,135],"payload":[43,137],"through":[44],"aggregation":[46,57,110],"entropy":[48],"coding.":[49],"However,":[50],"continuous-feature":[51],"coding":[52],"remains":[53],"costly,":[54],"query-agnostic":[56],"may":[58],"discard":[59],"task-relevant":[60],"local":[61],"evidence.":[62],"We":[63],"propose":[64],"query-guided":[65],"task-oriented":[66],"(Q-TOFC)":[69],"for":[70],"device-edge":[71],"inference.":[73],"Q-TOFC":[74,133],"employs":[75],"residual":[76],"vector":[77],"quantization":[78,114],"(RVQ)":[79],"encode":[81],"each":[82],"merged":[83],"as":[85],"a":[86,113],"compact":[87],"sequence":[88],"of":[89],"codebook":[90],"indices,":[91],"reducing":[92],"its":[93],"representation":[94],"cost":[95],"allowing":[97],"more":[98],"features":[99],"be":[101],"transmitted.":[102],"It":[103],"further":[104,153],"incorporates":[105],"query":[106],"relevance":[107],"into":[108],"uses":[112],"error":[115],"compensation":[116],"adapter":[117],"mitigate":[119],"distortion":[121],"introduced":[122],"by":[123,138],"discrete":[124],"quantization.":[125],"Experiments":[126],"seven":[128],"benchmarks":[130],"show":[131],"that":[132],"53.6%":[139],"relative":[140],"TOFC":[142],"while":[143],"maintaining":[144],"comparable":[145],"average":[146],"normalized":[147],"task":[148],"performance.":[149],"End-to-end":[150],"latency":[151,156],"evaluations":[152],"demonstrate":[154],"lower":[155],"under":[157],"bandwidth-constrained":[158],"uplinks.":[159]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-01T00:00:00"}