{"id":"https://openalex.org/W4415308626","doi":"https://doi.org/10.1145/3725843.3762817","title":"StreamTensor: Make Tensors Stream in Dataflow Accelerators for LLMs","display_name":"StreamTensor: Make Tensors Stream in Dataflow Accelerators for LLMs","publication_year":2025,"publication_date":"2025-10-17","ids":{"openalex":"https://openalex.org/W4415308626","doi":"https://doi.org/10.1145/3725843.3762817"},"language":"en","primary_location":{"id":"doi:10.1145/3725843.3762817","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3725843.3762817","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2509.13694","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5088715551","display_name":"Hanchen Ye","orcid":"https://orcid.org/0000-0002-6646-8146"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Hanchen Ye","raw_affiliation_strings":["University of Illinois Urbana-Champaign, Urbana, Illinois, USA"],"affiliations":[{"raw_affiliation_string":"University of Illinois Urbana-Champaign, Urbana, Illinois, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056321228","display_name":"Deming Chen","orcid":"https://orcid.org/0000-0002-3016-0270"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Deming Chen","raw_affiliation_strings":["Inspirit IoT, Inc., Champaign, Illinois, USA and University of Illinois Urbana-Champaign, Urbana, Illinois, USA"],"affiliations":[{"raw_affiliation_string":"Inspirit IoT, Inc., Champaign, Illinois, USA and University of Illinois Urbana-Champaign, Urbana, Illinois, USA","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5088715551"],"corresponding_institution_ids":["https://openalex.org/I157725225"],"apc_list":null,"apc_paid":null,"fwci":2.1983,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.88861335,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"201","last_page":"216"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9991999864578247,"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":0.9991999864578247,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.9983999729156494,"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/T12810","display_name":"Real-time simulation and control systems","score":0.9944999814033508,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems 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/dataflow","display_name":"Dataflow","score":0.9279999732971191},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.5792999863624573},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5580999851226807},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5167999863624573},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5105000138282776},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.49880000948905945},{"id":"https://openalex.org/keywords/dataflow-architecture","display_name":"Dataflow architecture","score":0.4812999963760376},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.43790000677108765},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.41780000925064087}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.9279999732971191},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8536999821662903},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.5792999863624573},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5580999851226807},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5460000038146973},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5167999863624573},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5105000138282776},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.49880000948905945},{"id":"https://openalex.org/C176727019","wikidata":"https://www.wikidata.org/wiki/Q1172415","display_name":"Dataflow architecture","level":3,"score":0.4812999963760376},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.43790000677108765},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.41780000925064087},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.40369999408721924},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.39879998564720154},{"id":"https://openalex.org/C107027933","wikidata":"https://www.wikidata.org/wiki/Q2006448","display_name":"Stream processing","level":2,"score":0.3756999969482422},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.3418000042438507},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.31690001487731934},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.301800012588501},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2816999852657318},{"id":"https://openalex.org/C184596265","wikidata":"https://www.wikidata.org/wiki/Q2651576","display_name":"Model of computation","level":3,"score":0.2791000008583069},{"id":"https://openalex.org/C200833197","wikidata":"https://www.wikidata.org/wiki/Q333707","display_name":"Compile time","level":3,"score":0.2768000066280365},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2745000123977661},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.2685999870300293},{"id":"https://openalex.org/C51290061","wikidata":"https://www.wikidata.org/wiki/Q1936765","display_name":"Uniform memory access","level":4,"score":0.26019999384880066},{"id":"https://openalex.org/C70388272","wikidata":"https://www.wikidata.org/wiki/Q5968558","display_name":"IBM","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.2581000030040741},{"id":"https://openalex.org/C489000","wikidata":"https://www.wikidata.org/wiki/Q747385","display_name":"Data flow diagram","level":2,"score":0.2554999887943268},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.25450000166893005}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3725843.3762817","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3725843.3762817","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2509.13694","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2509.13694","pdf_url":"https://arxiv.org/pdf/2509.13694","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2509.13694","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2509.13694","pdf_url":"https://arxiv.org/pdf/2509.13694","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":[],"awards":[{"id":"https://openalex.org/G539115378","display_name":null,"funder_award_id":"2141064","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":56,"referenced_works":["https://openalex.org/W1568192366","https://openalex.org/W1867684417","https://openalex.org/W1983394510","https://openalex.org/W1991801986","https://openalex.org/W2070385897","https://openalex.org/W2088899870","https://openalex.org/W2091158003","https://openalex.org/W2153658438","https://openalex.org/W2153911262","https://openalex.org/W2289252105","https://openalex.org/W2515406844","https://openalex.org/W2538051842","https://openalex.org/W2626953429","https://openalex.org/W2790678706","https://openalex.org/W2798482398","https://openalex.org/W2899691047","https://openalex.org/W2899915146","https://openalex.org/W2918037051","https://openalex.org/W2949676527","https://openalex.org/W2954698171","https://openalex.org/W2987684178","https://openalex.org/W3011239338","https://openalex.org/W3033664232","https://openalex.org/W3042501387","https://openalex.org/W3122286897","https://openalex.org/W3133395503","https://openalex.org/W3158233278","https://openalex.org/W3190681843","https://openalex.org/W3211485653","https://openalex.org/W3211800879","https://openalex.org/W4229485246","https://openalex.org/W4233532211","https://openalex.org/W4243997082","https://openalex.org/W4251637954","https://openalex.org/W4297097345","https://openalex.org/W4308083513","https://openalex.org/W4312121047","https://openalex.org/W4315588165","https://openalex.org/W4318541677","https://openalex.org/W4319870545","https://openalex.org/W4321637375","https://openalex.org/W4324292875","https://openalex.org/W4327930469","https://openalex.org/W4360831816","https://openalex.org/W4380874786","https://openalex.org/W4387187832","https://openalex.org/W4393406930","https://openalex.org/W4393577198","https://openalex.org/W4393949386","https://openalex.org/W4394661856","https://openalex.org/W4394871707","https://openalex.org/W4394998532","https://openalex.org/W4401331008","https://openalex.org/W4404954357","https://openalex.org/W4406531246","https://openalex.org/W6973789289"],"related_works":[],"abstract_inverted_index":{"Efficient":[0],"execution":[1],"of":[2],"deep":[3,153],"learning":[4,154],"workloads":[5],"on":[6,109,112],"dataflow":[7,57],"architectures":[8],"is":[9],"crucial":[10],"for":[11,150],"overcoming":[12],"memory":[13,36,79,100],"bottlenecks":[14],"and":[15,39,54,78,93,102,122,133,135],"maximizing":[16],"performance.":[17,107],"While":[18],"streaming":[19,104],"intermediate":[20],"results":[21],"between":[22],"computation":[23],"kernels":[24],"can":[25],"significantly":[26],"improve":[27],"efficiency,":[28,101],"existing":[29],"approaches":[30],"struggle":[31],"with":[32],"inter-kernel":[33],"correlations,":[34],"external":[35],"access":[37],"management,":[38],"buffer":[40,76],"optimization.":[41,80],"In":[42],"this":[43],"work,":[44],"we":[45],"propose":[46],"StreamTensor,":[47],"a":[48,61,147],"compiler":[49],"framework":[50],"that":[51],"automatically":[52],"constructs":[53],"optimizes":[55],"stream-based":[56],"accelerators.":[58],"StreamTensor":[59,96,117],"introduces":[60],"novel":[62],"iterative":[63],"tensor":[64,89],"type":[65],"system":[66],"to":[67,105,120,127,137,143],"explicitly":[68],"encode":[69],"stream":[70],"layouts,":[71],"enabling":[72],"seamless":[73],"kernel":[74,91],"fusion,":[75,92],"allocation,":[77,95],"By":[81],"systematically":[82],"exploring":[83],"three":[84],"hierarchical":[85],"design":[86],"spaces,":[87],"including":[88],"tiling,":[90],"resource":[94],"balances":[97],"computational":[98],"intensity,":[99],"data":[103],"maximize":[106],"Based":[108],"FPGA":[110,130],"evaluations":[111],"Large":[113],"Language":[114],"Models":[115],"(LLM),":[116],"achieves":[118],"up":[119,136],"0.76x":[121],"0.64x":[123],"lower":[124],"latency":[125],"compared":[126,142],"the":[128],"state-of-the-art":[129],"LLM":[131],"accelerators":[132],"GPUs,":[134,144],"1.99x":[138],"higher":[139],"energy":[140],"efficiency":[141],"making":[145],"it":[146],"promising":[148],"approach":[149],"scalable":[151],"dataflow-based":[152],"acceleration.":[155]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-18T00:00:00"}