{"id":"https://openalex.org/W4413014944","doi":"https://doi.org/10.1145/3676642.3736114","title":"<scp>Neuralink:</scp> Fast on-Device LLM Inference with Neuron Co-Activation Linking","display_name":"<scp>Neuralink:</scp> Fast on-Device LLM Inference with Neuron Co-Activation Linking","publication_year":2025,"publication_date":"2025-08-06","ids":{"openalex":"https://openalex.org/W4413014944","doi":"https://doi.org/10.1145/3676642.3736114"},"language":"en","primary_location":{"id":"doi:10.1145/3676642.3736114","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676642.3736114","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676642.3736114","source":null,"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3676642.3736114","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5067496010","display_name":"Tuowei Wang","orcid":"https://orcid.org/0009-0006-8272-8151"},"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":true,"raw_author_name":"Tuowei Wang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103639754","display_name":"Rong Fan","orcid":"https://orcid.org/0009-0009-3305-4631"},"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":"Ruwen Fan","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014444241","display_name":"Minxing Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Minxing Huang","raw_affiliation_strings":["Tianjin University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093259788","display_name":"Zixu Hao","orcid":"https://orcid.org/0009-0007-1671-1367"},"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":"Zixu Hao","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100377529","display_name":"Kun Li","orcid":"https://orcid.org/0000-0002-1013-1325"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kun Li","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101534161","display_name":"Ting Cao","orcid":"https://orcid.org/0000-0002-9107-013X"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ting Cao","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052095173","display_name":"Youyou Lu","orcid":"https://orcid.org/0000-0002-6214-5390"},"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":"Youyou Lu","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000488023","display_name":"Yaoxue Zhang","orcid":"https://orcid.org/0000-0002-4833-9789"},"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":"Yaoxue Zhang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5015419107","display_name":"Ju Ren","orcid":"https://orcid.org/0000-0003-2782-183X"},"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":"Ju Ren","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5067496010"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":1.2512,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.82325184,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"147","last_page":"162"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.989799976348877,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.989799976348877,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9861999750137329,"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/T10320","display_name":"Neural Networks and Applications","score":0.9850000143051147,"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/inference","display_name":"Inference","score":0.619497537612915},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5792732834815979},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3182552456855774}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.619497537612915},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5792732834815979},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3182552456855774}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3676642.3736114","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676642.3736114","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676642.3736114","source":null,"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3676642.3736114","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676642.3736114","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676642.3736114","source":null,"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4413014944.pdf","grobid_xml":"https://content.openalex.org/works/W4413014944.grobid-xml"},"referenced_works_count":41,"referenced_works":["https://openalex.org/W1970691969","https://openalex.org/W1990832096","https://openalex.org/W1994584977","https://openalex.org/W2022132320","https://openalex.org/W2061344791","https://openalex.org/W2120972216","https://openalex.org/W2126004407","https://openalex.org/W2138019504","https://openalex.org/W2606555609","https://openalex.org/W2626696598","https://openalex.org/W2963159690","https://openalex.org/W2963382930","https://openalex.org/W2998617917","https://openalex.org/W3034628778","https://openalex.org/W3083410900","https://openalex.org/W3090912412","https://openalex.org/W3130660608","https://openalex.org/W3137147200","https://openalex.org/W3184606595","https://openalex.org/W4205983429","https://openalex.org/W4236965008","https://openalex.org/W4298422451","https://openalex.org/W4309591680","https://openalex.org/W4319662928","https://openalex.org/W4366489368","https://openalex.org/W4387302777","https://openalex.org/W4387321086","https://openalex.org/W4387995158","https://openalex.org/W4389364428","https://openalex.org/W4392223155","https://openalex.org/W4393371888","https://openalex.org/W4399121418","https://openalex.org/W4404783432","https://openalex.org/W4405755563","https://openalex.org/W4405756292","https://openalex.org/W4406603752","https://openalex.org/W4409328045","https://openalex.org/W6615861906","https://openalex.org/W6724998850","https://openalex.org/W6727099177","https://openalex.org/W6778883912"],"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":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4,31],"achieved":[5],"remarkable":[6],"success":[7],"across":[8],"various":[9],"domains,":[10],"yet":[11],"deploying":[12],"them":[13],"on":[14,79,97,141,163,174],"mobile":[15,36],"devices":[16],"remains":[17],"an":[18,133,145],"arduous":[19],"challenge":[20],"due":[21],"to":[22,34,56,119,156,182,188],"their":[23],"extensive":[24],"computational":[25],"and":[26,124,144,153,168],"memory":[27],"demands.While":[28],"lightweight":[29],"LLMs":[30,169],"been":[32],"developed":[33],"fit":[35],"environments,":[37],"they":[38],"suffer":[39],"from":[40],"degraded":[41],"model":[42,62],"accuracy.In":[43],"contrast,":[44],"sparsitybased":[45],"techniques":[46],"minimize":[47],"DRAM":[48,57],"usage":[49],"by":[50,74,99],"selectively":[51],"transferring":[52],"only":[53],"relevant":[54],"neurons":[55,113],"while":[58],"retaining":[59],"the":[60,107,183,185],"full":[61],"in":[63,103,178],"external":[64],"storage,":[65],"such":[66,69],"as":[67],"flash.However,":[68],"approaches":[70],"are":[71,117],"critically":[72],"limited":[73],"numerous":[75],"I/O":[76,126],"operations,":[77],"particularly":[78],"smartphones":[80,98,167],"with":[81,159],"severe":[82],"IOPS":[83],"constraints.In":[84],"this":[85],"paper,":[86],"we":[87],"propose":[88],"Neuralink,":[89],"a":[90,130,164,196],"novel":[91],"approach":[92,128],"that":[93,136,148,171],"accelerates":[94],"LLM":[95],"inference":[96],"optimizing":[100],"neuron":[101,138],"placement":[102,139,191],"flash":[104],"memory.Neuralink":[105],"leverages":[106],"concept":[108],"of":[109,166],"Neuron":[110],"Co-Activation,":[111],"where":[112],"frequently":[114],"activated":[115],"together":[116],"linked":[118],"facilitate":[120],"continuous":[121],"read":[122],"access":[123,152],"optimize":[125,189],"efficiency.Our":[127],"incorporates":[129],"two-stage":[131],"solution:":[132],"offline":[134],"stage":[135,147],"reorganizes":[137],"based":[140],"co-activation":[142],"patterns,":[143],"online":[146],"employs":[149],"tailored":[150],"data":[151],"caching":[154],"strategies":[155],"align":[157],"well":[158],"hardware":[160],"characteristics.Evaluations":[161],"conducted":[162],"variety":[165],"demonstrate":[170],"Neuralink":[172,194],"achieves":[173],"average":[175],"1.49":[176],"improvements":[177],"end-to-end":[179],"latency":[180],"compared":[181],"state-of-the-art.As":[184],"first":[186],"solution":[187],"storage":[190],"under":[192],"sparsity,":[193],"explores":[195],"new":[197]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-02T15:55:50.835912","created_date":"2025-08-06T00:00:00"}