{"id":"https://openalex.org/W7131115217","doi":"https://doi.org/10.1109/cgo68049.2026.11395195","title":"Eliminating Redundancy: Ultra-compact Code Generation for Programmable Dataflow Accelerators","display_name":"Eliminating Redundancy: Ultra-compact Code Generation for Programmable Dataflow Accelerators","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7131115217","doi":"https://doi.org/10.1109/cgo68049.2026.11395195"},"language":null,"primary_location":{"id":"doi:10.1109/cgo68049.2026.11395195","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cgo68049.2026.11395195","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5000520819","display_name":"Prasanth Chatarasi","orcid":"https://orcid.org/0000-0002-0974-4001"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]},{"id":"https://openalex.org/I4210106712","display_name":"Alliance for Safe Kids","ror":"https://ror.org/01922pt58","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210106712"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Prasanth Chatarasi","raw_affiliation_strings":["IBM Research, Yorktown Heights,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights,USA","institution_ids":["https://openalex.org/I4210106712","https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126600147","display_name":"Alex Gatea","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113654","display_name":"IBM (Canada)","ror":"https://ror.org/025sxka56","country_code":"CA","type":"company","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210113654"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Alex Gatea","raw_affiliation_strings":["IBM Software,Toronto,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Software,Toronto,Canada","institution_ids":["https://openalex.org/I4210113654"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012183381","display_name":"Bardia Mahjour","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113654","display_name":"IBM (Canada)","ror":"https://ror.org/025sxka56","country_code":"CA","type":"company","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210113654"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Bardia Mahjour","raw_affiliation_strings":["IBM Software,Toronto,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Software,Toronto,Canada","institution_ids":["https://openalex.org/I4210113654"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126639336","display_name":"Jintao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jintao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008540803","display_name":"Alberto Mannari","orcid":null},"institutions":[{"id":"https://openalex.org/I4210126328","display_name":"IBM Research - Zurich","ror":"https://ror.org/02js37d36","country_code":"CH","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210126328"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Alberto Mannari","raw_affiliation_strings":["IBM Research,Zurich,Switzerland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research,Zurich,Switzerland","institution_ids":["https://openalex.org/I4210126328"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126588675","display_name":"Chris Bowler","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113654","display_name":"IBM (Canada)","ror":"https://ror.org/025sxka56","country_code":"CA","type":"company","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210113654"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Chris Bowler","raw_affiliation_strings":["IBM Software,Toronto,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Software,Toronto,Canada","institution_ids":["https://openalex.org/I4210113654"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119849951","display_name":"Shubham Jain","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]},{"id":"https://openalex.org/I4210106712","display_name":"Alliance for Safe Kids","ror":"https://ror.org/01922pt58","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210106712"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shubham Jain","raw_affiliation_strings":["IBM Research, Yorktown Heights,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights,USA","institution_ids":["https://openalex.org/I4210106712","https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064508687","display_name":"Masoud Ataei Jaliseh","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113654","display_name":"IBM (Canada)","ror":"https://ror.org/025sxka56","country_code":"CA","type":"company","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210113654"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Masoud Ataei Jaliseh","raw_affiliation_strings":["IBM Software,Toronto,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Software,Toronto,Canada","institution_ids":["https://openalex.org/I4210113654"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126597297","display_name":"Nicole Khoun","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113654","display_name":"IBM (Canada)","ror":"https://ror.org/025sxka56","country_code":"CA","type":"company","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210113654"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Nicole Khoun","raw_affiliation_strings":["IBM Software,Toronto,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Software,Toronto,Canada","institution_ids":["https://openalex.org/I4210113654"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126617136","display_name":"Kamlesh Kumar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kamlesh Kumar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104063917","display_name":"Viji Srinivasan","orcid":null},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]},{"id":"https://openalex.org/I4210106712","display_name":"Alliance for Safe Kids","ror":"https://ror.org/01922pt58","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210106712"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Viji Srinivasan","raw_affiliation_strings":["IBM Research, Yorktown Heights,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights,USA","institution_ids":["https://openalex.org/I4210106712","https://openalex.org/I1341412227"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5010094713","display_name":"Swagath Venkataramani","orcid":"https://orcid.org/0000-0002-0470-6364"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]},{"id":"https://openalex.org/I4210106712","display_name":"Alliance for Safe Kids","ror":"https://ror.org/01922pt58","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210106712"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Swagath Venkataramani","raw_affiliation_strings":["IBM Research, Yorktown Heights,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights,USA","institution_ids":["https://openalex.org/I4210106712","https://openalex.org/I1341412227"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"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":"44","last_page":"56"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.4952000081539154,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.4952000081539154,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.2167000025510788,"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/T10126","display_name":"Logic, programming, and type systems","score":0.05009999871253967,"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/dataflow","display_name":"Dataflow","score":0.8389000296592712},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.7670999765396118},{"id":"https://openalex.org/keywords/code-generation","display_name":"Code generation","score":0.647599995136261},{"id":"https://openalex.org/keywords/operand","display_name":"Operand","score":0.5849999785423279},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.48420000076293945},{"id":"https://openalex.org/keywords/loop","display_name":"Loop (graph theory)","score":0.4023999869823456},{"id":"https://openalex.org/keywords/nested-loop-join","display_name":"Nested loop join","score":0.39809998869895935},{"id":"https://openalex.org/keywords/critical-path-method","display_name":"Critical path method","score":0.3946000039577484},{"id":"https://openalex.org/keywords/instruction-set","display_name":"Instruction set","score":0.3492000102996826}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.8389000296592712},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8348000049591064},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.7670999765396118},{"id":"https://openalex.org/C133162039","wikidata":"https://www.wikidata.org/wiki/Q1061077","display_name":"Code generation","level":3,"score":0.647599995136261},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6187000274658203},{"id":"https://openalex.org/C55526617","wikidata":"https://www.wikidata.org/wiki/Q719375","display_name":"Operand","level":2,"score":0.5849999785423279},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.48420000076293945},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.42669999599456787},{"id":"https://openalex.org/C184670325","wikidata":"https://www.wikidata.org/wiki/Q512604","display_name":"Loop (graph theory)","level":2,"score":0.4023999869823456},{"id":"https://openalex.org/C1306188","wikidata":"https://www.wikidata.org/wiki/Q4060687","display_name":"Nested loop join","level":2,"score":0.39809998869895935},{"id":"https://openalex.org/C115874739","wikidata":"https://www.wikidata.org/wiki/Q825377","display_name":"Critical path method","level":2,"score":0.3946000039577484},{"id":"https://openalex.org/C202491316","wikidata":"https://www.wikidata.org/wiki/Q272683","display_name":"Instruction set","level":2,"score":0.3492000102996826},{"id":"https://openalex.org/C128916667","wikidata":"https://www.wikidata.org/wiki/Q1343660","display_name":"Register allocation","level":3,"score":0.3253999948501587},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.32330000400543213},{"id":"https://openalex.org/C134835016","wikidata":"https://www.wikidata.org/wiki/Q690265","display_name":"Lookup table","level":2,"score":0.2921999990940094},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.28220000863075256},{"id":"https://openalex.org/C489000","wikidata":"https://www.wikidata.org/wiki/Q747385","display_name":"Data flow diagram","level":2,"score":0.27889999747276306},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.27709999680519104},{"id":"https://openalex.org/C190902152","wikidata":"https://www.wikidata.org/wiki/Q1325106","display_name":"Optimizing compiler","level":3,"score":0.2750999927520752},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.2705000042915344},{"id":"https://openalex.org/C2779624466","wikidata":"https://www.wikidata.org/wiki/Q838119","display_name":"For loop","level":3,"score":0.25600001215934753},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.2524000108242035},{"id":"https://openalex.org/C88468194","wikidata":"https://www.wikidata.org/wiki/Q1172416","display_name":"Data-flow analysis","level":3,"score":0.25040000677108765},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cgo68049.2026.11395195","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cgo68049.2026.11395195","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.7337673306465149,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1485882819","https://openalex.org/W1519455305","https://openalex.org/W1550771019","https://openalex.org/W1969031936","https://openalex.org/W1982106660","https://openalex.org/W1993402810","https://openalex.org/W2000722076","https://openalex.org/W2040455459","https://openalex.org/W2115895988","https://openalex.org/W2154790323","https://openalex.org/W2291192259","https://openalex.org/W2340812064","https://openalex.org/W2790925711","https://openalex.org/W2898985762","https://openalex.org/W2906043559","https://openalex.org/W2918037051","https://openalex.org/W2945146780","https://openalex.org/W2963673357","https://openalex.org/W2979310060","https://openalex.org/W2980104813","https://openalex.org/W2981680926","https://openalex.org/W3122286897","https://openalex.org/W3187235458","https://openalex.org/W3190062760","https://openalex.org/W4220662607","https://openalex.org/W4220706374","https://openalex.org/W4244024631","https://openalex.org/W4246434167","https://openalex.org/W4281806264","https://openalex.org/W4379536406","https://openalex.org/W4380881154","https://openalex.org/W4402475869"],"related_works":[],"abstract_inverted_index":{"Modern":[0],"AI":[1],"accelerators":[2],"adopt":[3],"dataflow":[4],"architectures":[5],"to":[6,219],"achieve":[7],"both":[8],"high":[9],"peak":[10],"throughput":[11],"(TOPS)":[12],"and":[13,22,35,54,60,71,123,166,175,182,194],"energy":[14,59],"efficiency":[15],"(TOPS/W).":[16],"These":[17],"designs":[18],"feature":[19],"wide":[20],"datapaths":[21],"hierarchical":[23,169],"scratchpad":[24],"memories":[25],"that":[26,48,98,214],"supply":[27],"dense":[28],"compute":[29],"arrays":[30],"with":[31,150],"high-bandwidth":[32],"data":[33,50],"access":[34],"extensive":[36],"operand":[37],"reuse.":[38],"Complementing":[39],"the":[40,88,103,109,178,210],"compute\u2013memory":[41],"subsystem":[42],"is":[43],"a":[44,84,141,154,203],"lightweight":[45],"control":[46],"path":[47],"orchestrates":[49],"movement,":[51],"program":[52],"loading,":[53],"register":[55],"initialization.":[56],"To":[57],"reduce":[58],"area":[61],"overheads,":[62],"conventional":[63],"processor":[64],"features\u2014such":[65],"as":[66],"instruction":[67,96,105],"caches,":[68],"execution":[69],"stacks,":[70],"branch":[72],"speculation\u2014are":[73],"deliberately":[74],"omitted.":[75],"While":[76],"this":[77,113],"streamlined":[78],"design":[79],"maximizes":[80],"efficiency,":[81],"it":[82],"shifts":[83],"critical":[85],"responsibility":[86],"onto":[87],"compiler:":[89],"transforming":[90],"complex":[91],"kernels":[92],"into":[93,140,153],"highly":[94],"compact":[95,131,168],"streams":[97],"must":[99],"fit":[100],"entirely":[101],"within":[102],"limited":[104],"buffers":[106],"(IBUFFs)":[107],"of":[108,207],"accelerator\u2019s":[110],"programmable":[111],"units.In":[112],"paper,":[114],"we":[115],"introduce":[116],"two":[117],"novel":[118],"compiler":[119,181],"transformations\u2014Loop":[120],"Absorption":[121,135],"(LA)":[122],"Loop":[124,134],"Index":[125],"Set":[126],"Merging":[127],"(LISM)":[128],"for":[129],"ultra":[130],"code":[132,164],"generation.":[133],"merges":[136],"isomorphic":[137],"sibling":[138],"operations":[139],"single":[142],"loop":[143,170],"body,":[144],"while":[145],"LISM":[146,176],"unifies":[147],"adjacent":[148],"loops":[149],"similar":[151],"bodies":[152],"unified":[155],"iteration":[156],"space.":[157],"Together,":[158],"these":[159,197],"complementary":[160],"techniques":[161],"eliminate":[162],"redundant":[163],"patterns":[165],"produce":[167],"nests.":[171],"We":[172],"implement":[173],"LA":[174],"in":[177],"IBM":[179],"Spyre":[180],"evaluate":[183],"them":[184],"on":[185],"diverse":[186],"deep":[187],"learning":[188],"workloads":[189],"including":[190],"ResNet-50,":[191],"Inception-v3,":[192],"SSD,":[193],"BERT-Large.":[195],"Across":[196],"models,":[198],"our":[199],"combined":[200],"approach":[201],"achieves":[202],"geometric":[204],"mean":[205],"compression":[206],"1.48\u00d7":[208],"over":[209],"baseline,":[211],"enabling":[212],"layers":[213],"previously":[215],"exceeded":[216],"IBUFF":[217],"capacity":[218],"compile":[220],"successfully.":[221]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-02-24T00:00:00"}