{"id":"https://openalex.org/W7161279463","doi":"https://doi.org/10.48550/arxiv.2605.13887","title":"Breaking Global Self-Attention Bottlenecks in Transformer-based Spiking Neural Networks with Local Structure-Aware Self-Attention","display_name":"Breaking Global Self-Attention Bottlenecks in Transformer-based Spiking Neural Networks with Local Structure-Aware Self-Attention","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161279463","doi":"https://doi.org/10.48550/arxiv.2605.13887"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.13887","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.13887","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.13887","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136202253","display_name":"Lingdong Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Lingdong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136268036","display_name":"Hangming Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Hangming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136208775","display_name":"Qiang Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Qiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9319999814033508,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9319999814033508,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.02800000086426735,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.025299999862909317,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/spiking-neural-network","display_name":"Spiking neural network","score":0.8483999967575073},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.8166000247001648},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4855000078678131},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4618000090122223},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.4406000077724457},{"id":"https://openalex.org/keywords/neuromorphic-engineering","display_name":"Neuromorphic engineering","score":0.42890000343322754},{"id":"https://openalex.org/keywords/random-neural-network","display_name":"Random neural network","score":0.4138999879360199},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.35929998755455017}],"concepts":[{"id":"https://openalex.org/C11731999","wikidata":"https://www.wikidata.org/wiki/Q9067355","display_name":"Spiking neural network","level":3,"score":0.8483999967575073},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.8166000247001648},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7821000218391418},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5803999900817871},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4855000078678131},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.45170000195503235},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.4406000077724457},{"id":"https://openalex.org/C151927369","wikidata":"https://www.wikidata.org/wiki/Q1981312","display_name":"Neuromorphic engineering","level":3,"score":0.42890000343322754},{"id":"https://openalex.org/C33010914","wikidata":"https://www.wikidata.org/wiki/Q7291980","display_name":"Random neural network","level":4,"score":0.4138999879360199},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.35929998755455017},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.35899999737739563},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.35510000586509705},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.29120001196861267},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C123138037","wikidata":"https://www.wikidata.org/wiki/Q5570871","display_name":"Global network","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C2777655017","wikidata":"https://www.wikidata.org/wiki/Q1501161","display_name":"Toolbox","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C194544171","wikidata":"https://www.wikidata.org/wiki/Q21105679","display_name":"Gating","level":2,"score":0.26109999418258667},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.25220000743865967}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.13887","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.13887","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.13887","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.13887","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":"article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.8953967094421387,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Transformer-based":[0,17,95,144],"Spiking":[1,90,96,101,108],"Neural":[2,97],"Networks":[3],"(SNNs)":[4],"integrate":[5],"SNNs":[6,18],"with":[7,74],"global":[8,57,61],"self-attention":[9,58],"and":[10,47,68,77,105,128,154,165],"have":[11],"demonstrated":[12],"impressive":[13],"performance.":[14],"However,":[15],"existing":[16,142],"suffer":[19],"from":[20],"two":[21],"fundamental":[22],"limitations.":[23],"First,":[24],"they":[25],"typically":[26],"employ":[27],"max":[28,40],"pooling":[29,41],"layers":[30],"to":[31,49,123,141,179],"reduce":[32],"the":[33,39,44,56,75,112,148,175],"size":[34],"of":[35,80,177],"feature":[36,62],"maps,":[37],"but":[38],"captures":[42],"only":[43],"strongest":[45],"response":[46],"fails":[48],"comprehensively":[50],"preserve":[51],"representative":[52],"regional":[53],"features.":[54],"Second,":[55],"involves":[59],"all":[60],"interactions,":[63],"resulting":[64],"in":[65,167,187],"computational":[66,70],"redundancy":[67],"quadratic":[69],"complexity,":[71],"thus":[72],"conflicting":[73],"sparse":[76],"energy-efficient":[78,181],"characteristics":[79],"SNNs.":[81,145],"To":[82],"address":[83],"these":[84],"challenges,":[85],"we":[86],"develop":[87],"Local":[88,106],"Structure-Aware":[89,107],"Transformer":[91],"(LSFormer),":[92],"a":[93,118],"novel":[94],"Network":[98],"that":[99,134],"incorporates":[100],"Response":[102],"Pooling":[103],"(SPooling)":[104],"Self-Attention":[109],"(LS-SSA).":[110],"For":[111],"first":[113],"time,":[114],"our":[115,135],"LSFormer":[116,136,158,178],"leverages":[117],"local":[119,126],"dilated":[120],"window":[121],"mechanism":[122],"capture":[124],"both":[125],"details":[127],"long-range":[129],"dependencies.":[130],"Experimental":[131],"results":[132,173],"demonstrate":[133],"achieves":[137],"state-of-the-art":[138,161],"performance":[139],"compared":[140],"advanced":[143],"Notably,":[146],"on":[147],"more":[149],"challenging":[150],"static":[151],"dataset":[152,156],"Tiny-ImageNet":[153],"neuromorphic":[155],"N-CALTECH101,":[157],"substantially":[159],"outperforms":[160],"baselines":[162],"by":[163],"4.3\\%":[164],"8.6\\%":[166],"top-1":[168],"classification":[169],"accuracy,":[170],"respectively.":[171],"These":[172],"highlight":[174],"potential":[176],"advance":[180],"spiking":[182],"models":[183],"toward":[184],"practical":[185],"deployment":[186],"large-scale":[188],"vision":[189],"applications.":[190]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-16T00:00:00"}