{"id":"https://openalex.org/W7160643953","doi":"https://doi.org/10.48550/arxiv.2605.05819","title":"HCInfer: An Efficient Inference System via Error Compensation for Resource-Constrained Devices","display_name":"HCInfer: An Efficient Inference System via Error Compensation for Resource-Constrained Devices","publication_year":2026,"publication_date":"2026-05-07","ids":{"openalex":"https://openalex.org/W7160643953","doi":"https://doi.org/10.48550/arxiv.2605.05819"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.05819","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05819","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.05819","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135638615","display_name":"Shen Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Shen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120051065","display_name":"Xiangwen Zhuge","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuge, Xiangwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135643081","display_name":"Zhe Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135645651","display_name":"Yingkun Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Yingkun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135649417","display_name":"Zheng Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Zheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135668662","display_name":"Yunhao Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yunhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.3276999890804291,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.3276999890804291,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.1858000010251999,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.10520000010728836,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/compensation","display_name":"Compensation (psychology)","score":0.6743000149726868},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.6618000268936157},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.6396999955177307},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6309999823570251},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5956000089645386},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.5819000005722046},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.5253000259399414},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5149999856948853}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7972000241279602},{"id":"https://openalex.org/C2780023022","wikidata":"https://www.wikidata.org/wiki/Q1338171","display_name":"Compensation (psychology)","level":2,"score":0.6743000149726868},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6618000268936157},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.6396999955177307},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6309999823570251},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5956000089645386},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5819000005722046},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.5253000259399414},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5149999856948853},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4975000023841858},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.4950000047683716},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3869999945163727},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3560999929904938},{"id":"https://openalex.org/C2779679103","wikidata":"https://www.wikidata.org/wiki/Q5251805","display_name":"Degradation (telecommunications)","level":2,"score":0.33899998664855957},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.3370000123977661},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.3334999978542328},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.31299999356269836},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.30809998512268066},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3018999993801117},{"id":"https://openalex.org/C81293917","wikidata":"https://www.wikidata.org/wiki/Q4189534","display_name":"System deployment","level":3,"score":0.29829999804496765},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.289000004529953},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.2831000089645386},{"id":"https://openalex.org/C103088060","wikidata":"https://www.wikidata.org/wiki/Q1062839","display_name":"Error detection and correction","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2531000077724457},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.2515999972820282}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.05819","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05819","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.05819","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05819","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"LLMs":[0],"often":[1,28],"struggle":[2],"with":[3],"memory-constrained":[4],"deployment":[5,25],"on":[6,100,135],"consumer-grade":[7],"hardware":[8],"due":[9],"to":[10,57,92,115,139,150],"their":[11],"massive":[12],"parameter":[13,62],"sizes.":[14],"While":[15],"existing":[16],"solutions":[17],"such":[18],"as":[19],"model":[20,141],"compression":[21,140],"and":[22,48,103,110,119,142],"offloading":[23],"improve":[24],"feasibility,":[26],"they":[27,59],"suffer":[29],"from":[30],"substantial":[31,61],"accuracy":[32,43,121,131],"degradation":[33],"or":[34],"severe":[35],"throughput":[36],"bottlenecks.":[37],"Recent":[38],"error":[39],"compensation":[40,71,91,108,117],"methods":[41],"recover":[42],"through":[44],"auxiliary":[45],"LoRA-style":[46],"branches,":[47],"we":[49,81],"observe":[50],"that":[51,88,126],"these":[52],"branches":[53],"are":[54],"inherently":[55],"amenable":[56],"offloading:":[58],"require":[60],"storage":[63],"but":[64],"access":[65],"only":[66],"a":[67,84,129,144],"small":[68],"subset":[69],"of":[70,133,147],"parameters":[72],"during":[73],"each":[74],"inference":[75,86],"step.":[76],"Motivated":[77],"by":[78],"this":[79],"opportunity,":[80],"propose":[82],"HCInfer,":[83],"heterogeneous":[85],"system":[87],"offloads":[89],"residual":[90],"the":[93,97,101],"CPU":[94],"while":[95],"executing":[96],"compressed":[98],"backbone":[99],"GPU,":[102],"further":[104],"introduces":[105],"an":[106],"asynchronous":[107],"pipeline":[109],"sensitivity-aware":[111],"dynamic":[112],"rank":[113],"allocation":[114],"hide":[116],"overhead":[118],"maximize":[120],"recovery.":[122],"Experimental":[123],"results":[124],"show":[125],"HCInfer":[127],"achieves":[128],"maximum":[130,145],"improvement":[132],"5.2%":[134],"downstream":[136],"tasks":[137],"compared":[138,149],"sustaining":[143],"speedup":[146],"10.4x":[148],"full-precision":[151],"model.":[152]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-09T00:00:00"}