{"id":"https://openalex.org/W7160523127","doi":"https://doi.org/10.1109/tvlsi.2026.3688907","title":"TriCIM: A General CIM-Capacity-Aware Framework for Optimizing Model, Layer, and Tile-Stationary Dataflows in CIM Accelerators","display_name":"TriCIM: A General CIM-Capacity-Aware Framework for Optimizing Model, Layer, and Tile-Stationary Dataflows in CIM Accelerators","publication_year":2026,"publication_date":"2026-05-07","ids":{"openalex":"https://openalex.org/W7160523127","doi":"https://doi.org/10.1109/tvlsi.2026.3688907"},"language":null,"primary_location":{"id":"doi:10.1109/tvlsi.2026.3688907","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2026.3688907","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5108047889","display_name":"Yi-Xiang Wang","orcid":"https://orcid.org/0000-0001-5697-0717"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jin Wang","raw_affiliation_strings":["School of Microelectronics, Southern University of Science and Technology, Shenzhen, Guangdong, China"],"raw_orcid":"https://orcid.org/0009-0003-8174-3542","affiliations":[{"raw_affiliation_string":"School of Microelectronics, Southern University of Science and Technology, Shenzhen, Guangdong, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084697396","display_name":"Yufu Zhang","orcid":"https://orcid.org/0000-0003-0648-2058"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yufu Zhang","raw_affiliation_strings":["School of Microelectronics, Southern University of Science and Technology, Shenzhen, Guangdong, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Microelectronics, Southern University of Science and Technology, Shenzhen, Guangdong, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135604574","display_name":"Longyang Lin","orcid":"https://orcid.org/0000-0002-4702-737X"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Longyang Lin","raw_affiliation_strings":["State Key Laboratory of Quantum Functional Materials, School of Microelectronics, Southern University of Science and Technology, Shenzhen, Guangdong, China"],"raw_orcid":"https://orcid.org/0000-0002-4702-737X","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Quantum Functional Materials, School of Microelectronics, Southern University of Science and Technology, Shenzhen, Guangdong, China","institution_ids":["https://openalex.org/I3045169105"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I3045169105"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.56443007,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"34","issue":"7","first_page":"2187","last_page":"2200"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.16910000145435333,"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":0.16910000145435333,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.1453000009059906,"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"}},{"id":"https://openalex.org/T14347","display_name":"Big Data and Digital Economy","score":0.10170000046491623,"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/dataflow","display_name":"Dataflow","score":0.9718999862670898},{"id":"https://openalex.org/keywords/dataflow-architecture","display_name":"Dataflow architecture","score":0.6807000041007996},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.626800000667572},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5052000284194946},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.39959999918937683},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.3889000117778778},{"id":"https://openalex.org/keywords/model-of-computation","display_name":"Model of computation","score":0.3465000092983246},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.3441999852657318},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.34150001406669617}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.9718999862670898},{"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/C176727019","wikidata":"https://www.wikidata.org/wiki/Q1172415","display_name":"Dataflow architecture","level":3,"score":0.6807000041007996},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.626800000667572},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5920000076293945},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5052000284194946},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.43140000104904175},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4099000096321106},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.39959999918937683},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C184596265","wikidata":"https://www.wikidata.org/wiki/Q2651576","display_name":"Model of computation","level":3,"score":0.3465000092983246},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.3441999852657318},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.34150001406669617},{"id":"https://openalex.org/C2776221188","wikidata":"https://www.wikidata.org/wiki/Q21072556","display_name":"Design space exploration","level":2,"score":0.33340001106262207},{"id":"https://openalex.org/C88468194","wikidata":"https://www.wikidata.org/wiki/Q1172416","display_name":"Data-flow analysis","level":3,"score":0.3249000012874603},{"id":"https://openalex.org/C489000","wikidata":"https://www.wikidata.org/wiki/Q747385","display_name":"Data flow diagram","level":2,"score":0.3127000033855438},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.30869999527931213},{"id":"https://openalex.org/C64260653","wikidata":"https://www.wikidata.org/wiki/Q1194864","display_name":"Electronic design automation","level":2,"score":0.3003999888896942},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2992999851703644},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.2854999899864197},{"id":"https://openalex.org/C67953723","wikidata":"https://www.wikidata.org/wiki/Q192525","display_name":"Workstation","level":2,"score":0.2825999855995178},{"id":"https://openalex.org/C37135326","wikidata":"https://www.wikidata.org/wiki/Q931942","display_name":"Design flow","level":2,"score":0.2676999866962433},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.25760000944137573},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tvlsi.2026.3688907","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2026.3688907","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.906218945980072,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G6839350787","display_name":null,"funder_award_id":"62274081","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Compute-in-memory":[0],"(CIM)":[1],"has":[2],"emerged":[3],"as":[4,107],"a":[5,74,93,122,138],"promising":[6],"paradigm":[7],"for":[8,28],"accelerating":[9],"neural":[10],"networks":[11],"(NNs),":[12],"offering":[13],"high":[14],"parallelism":[15],"and":[16,44,53,66,101,114,153,161,173,186,193,203],"superior":[17],"energy":[18],"efficiency.":[19,196],"To":[20],"fully":[21],"leverage":[22],"the":[23,56,143,157,162],"potential":[24],"of":[25,58,96],"CIM":[26,49,59,83,87,132,144,159,201],"accelerators":[27,202],"deploying":[29],"NNs,":[30],"it":[31,70],"is":[32],"essential":[33],"to":[34,72,92,128,130,189,222,233],"carefully":[35],"optimize":[36],"their":[37,62,112],"mapping":[38,67],"dataflows,":[39,213],"which":[40,110],"determine":[41],"how":[42],"computations":[43],"data":[45],"are":[46,90],"scheduled":[47],"across":[48,81,199],"resources":[50],"over":[51],"space":[52,146],"time.":[54],"However,":[55],"diversity":[57],"architectures,":[60],"with":[61],"varying":[63],"hardware":[64],"parameters":[65],"constraints,":[68],"makes":[69],"challenging":[71],"develop":[73],"unified":[75],"framework":[76,126],"that":[77,141,208],"efficiently":[78],"optimizes":[79],"dataflows":[80],"different":[82],"systems.":[84],"Most":[85],"existing":[86],"dataflow":[88,99,124,145,167,195],"frameworks":[89],"tailored":[91,175],"limited":[94],"range":[95],"architectures":[97],"or":[98],"types,":[100],"often":[102],"neglect":[103],"CIM-specific":[104],"features":[105],"such":[106],"weight-update":[108,183],"scheduling,":[109,184],"limits":[111],"effectiveness":[113],"generality.":[115],"In":[116],"this":[117],"work,":[118],"we":[119,169],"propose":[120,174],"TriCIM,":[121],"general":[123],"optimization":[125,176],"designed":[127],"adapt":[129],"diverse":[131],"architectures.":[133],"We":[134],"begin":[135],"by":[136],"introducing":[137],"CIM-capacity-aware":[139],"formulation":[140],"classifies":[142],"into":[147],"three":[148],"distinct":[149],"regions,":[150],"model-stationary,":[151],"layer-stationary,":[152],"tile-stationary,":[154],"based":[155],"on":[156],"available":[158],"capacity":[160],"model":[163],"size.":[164],"For":[165],"each":[166],"region,":[168],"identify":[170],"specific":[171],"bottlenecks":[172,192],"strategies,":[177],"including":[178],"load":[179],"balancing,":[180],"layer":[181],"grouping,":[182],"tiling,":[185],"inter-tile":[187],"ordering,":[188],"address":[190],"region-specific":[191],"enhance":[194],"Our":[197],"evaluation":[198],"various":[200],"representative":[204],"NN":[205],"models":[206],"demonstrates":[207],"TriCIM":[209],"consistently":[210],"finds":[211],"optimal":[212],"achieving":[214],"<inline-formula":[215,223],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[216,224],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[217,225],"<tex-math":[218,226],"notation=\"LaTeX\">$1.1\\times":[219],"$</tex-math>":[220,228],"</inline-formula>":[221,229],"notation=\"LaTeX\">$13.2\\times":[227],"runtime":[230],"speedup":[231],"compared":[232],"state-of-the-art":[234],"frameworks.":[235]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-08T00:00:00"}