{"id":"https://openalex.org/W4393578753","doi":"https://doi.org/10.1145/3626202.3637562","title":"FlightLLM: Efficient Large Language Model Inference with a Complete Mapping Flow on FPGAs","display_name":"FlightLLM: Efficient Large Language Model Inference with a Complete Mapping Flow on FPGAs","publication_year":2024,"publication_date":"2024-04-01","ids":{"openalex":"https://openalex.org/W4393578753","doi":"https://doi.org/10.1145/3626202.3637562"},"language":"en","primary_location":{"id":"doi:10.1145/3626202.3637562","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3626202.3637562","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3626202.3637562","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2024 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","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/10.1145/3626202.3637562","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5026831784","display_name":"Shulin Zeng","orcid":"https://orcid.org/0000-0002-1030-3748"},"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":"Shulin Zeng","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/A5058051415","display_name":"J Liu","orcid":"https://orcid.org/0009-0003-8280-9072"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Liu","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015946486","display_name":"Guohao Dai","orcid":"https://orcid.org/0000-0003-0849-3252"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guohao Dai","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020244808","display_name":"Xinhao Yang","orcid":"https://orcid.org/0009-0001-9739-2930"},"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":"Xinhao Yang","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/A5083724554","display_name":"Tianyu Fu","orcid":"https://orcid.org/0000-0003-3508-1755"},"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":"Tianyu Fu","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/A5023007010","display_name":"Hongyi Wang","orcid":"https://orcid.org/0009-0008-7095-7963"},"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":"Hongyi 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/A5074195618","display_name":"Wenheng Ma","orcid":"https://orcid.org/0000-0003-2349-7286"},"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":"Wenheng Ma","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/A5000640882","display_name":"Hanbo Sun","orcid":"https://orcid.org/0000-0002-7875-2064"},"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":"Hanbo Sun","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/A5005872704","display_name":"Shiyao Li","orcid":"https://orcid.org/0009-0007-4755-6881"},"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":"Shiyao Li","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/A5101728676","display_name":"Zixiao Huang","orcid":"https://orcid.org/0009-0000-1273-2573"},"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":"Zixiao Huang","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/A5043844741","display_name":"Yadong Dai","orcid":"https://orcid.org/0009-0009-9193-8494"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yadong Dai","raw_affiliation_strings":["Infinigence-AI, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Infinigence-AI, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053894346","display_name":"Jintao Li","orcid":"https://orcid.org/0009-0006-8903-8276"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jintao Li","raw_affiliation_strings":["Infinigence-AI, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Infinigence-AI, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003892449","display_name":"Zehao Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zehao Wang","raw_affiliation_strings":["Infinigence-AI, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Infinigence-AI, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002311002","display_name":"R. Zhang","orcid":"https://orcid.org/0009-0006-3401-5237"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ruoyu Zhang","raw_affiliation_strings":["Infinigence-AI, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Infinigence-AI, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031548623","display_name":"Kairui Wen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kairui Wen","raw_affiliation_strings":["Infinigence-AI, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Infinigence-AI, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086217226","display_name":"Xuefei Ning","orcid":"https://orcid.org/0000-0003-2209-8312"},"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":"Xuefei Ning","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/A5100445061","display_name":"Yu Wang","orcid":"https://orcid.org/0000-0001-6108-5157"},"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":"Yu Wang","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":17,"corresponding_author_ids":["https://openalex.org/A5026831784"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":34.7002,"has_fulltext":true,"cited_by_count":103,"citation_normalized_percentile":{"value":0.99833147,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"223","last_page":"234"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9991000294685364,"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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9991000294685364,"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/T10028","display_name":"Topic Modeling","score":0.9973999857902527,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9919000267982483,"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/computer-science","display_name":"Computer science","score":0.7630889415740967},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7146406769752502},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.676008403301239},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.4973466694355011},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4436507821083069},{"id":"https://openalex.org/keywords/flow","display_name":"Flow (mathematics)","score":0.4218588173389435},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.37163245677948},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.37062951922416687},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.28015416860580444},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.22516795992851257},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.09686490893363953}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7630889415740967},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7146406769752502},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.676008403301239},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.4973466694355011},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4436507821083069},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.4218588173389435},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.37163245677948},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.37062951922416687},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28015416860580444},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.22516795992851257},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.09686490893363953},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3626202.3637562","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3626202.3637562","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3626202.3637562","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2024 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3626202.3637562","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3626202.3637562","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3626202.3637562","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2024 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.8899999856948853,"display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1799421016","display_name":null,"funder_award_id":"61832007","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2045798616","display_name":null,"funder_award_id":"62325405","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3317480652","display_name":null,"funder_award_id":"Science","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G391238517","display_name":null,"funder_award_id":", and","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5939423041","display_name":null,"funder_award_id":"Technology","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5994120800","display_name":null,"funder_award_id":"Natural","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G663016133","display_name":null,"funder_award_id":"62204164","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6803541034","display_name":null,"funder_award_id":"U21B2031","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6839492696","display_name":null,"funder_award_id":"U19B2019","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G763519461","display_name":null,"funder_award_id":"62325405","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"},{"id":"https://openalex.org/G8262591375","display_name":null,"funder_award_id":"62104128","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"},{"id":"https://openalex.org/F4320323817","display_name":"Universitas Brawijaya","ror":"https://ror.org/01wk3d929"},{"id":"https://openalex.org/F4320329777","display_name":"Beijing National Research Center For Information Science And Technology","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4393578753.pdf","grobid_xml":"https://content.openalex.org/works/W4393578753.grobid-xml"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W2616014673","https://openalex.org/W2904487346","https://openalex.org/W3012561096","https://openalex.org/W3034855459","https://openalex.org/W3103168911","https://openalex.org/W3130554079","https://openalex.org/W3130641740","https://openalex.org/W3189877953","https://openalex.org/W3206453033","https://openalex.org/W4200506235","https://openalex.org/W4211085526","https://openalex.org/W4211090839","https://openalex.org/W4224903119","https://openalex.org/W4281651027","https://openalex.org/W4297097408","https://openalex.org/W4308083526","https://openalex.org/W4318541578","https://openalex.org/W4321466385","https://openalex.org/W4360831803","https://openalex.org/W4380881077","https://openalex.org/W4382202798","https://openalex.org/W4386763770","https://openalex.org/W4387321091"],"related_works":["https://openalex.org/W2111241003","https://openalex.org/W4200391368","https://openalex.org/W2355315220","https://openalex.org/W2210979487","https://openalex.org/W2074043759","https://openalex.org/W2316202402","https://openalex.org/W2373535795","https://openalex.org/W1967938402","https://openalex.org/W2386041993","https://openalex.org/W1608572506"],"abstract_inverted_index":{"Transformer-based":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"have":[5],"made":[6],"a":[7,80,116,154],"significant":[8],"impact":[9],"on":[10,84,165,188],"various":[11],"domains.":[12],"However,":[13,44],"LLMs'":[14],"efficiency":[15,176,181],"suffers":[16],"from":[17],"both":[18],"heavy":[19],"computation":[20,95,128],"and":[21,28,41,47,67,96,110,177,195],"memory":[22,65,97,112,139],"overheads.":[23,70],"Compression":[24],"techniques":[25],"like":[26],"sparsification":[27],"quantization":[29],"are":[30],"commonly":[31],"used":[32],"to":[33,56,121,137,145,159],"mitigate":[34],"the":[35,57,94,161,166,198,213],"gap":[36],"between":[37],"LLM's":[38],"computation/memory":[39],"overheads":[40],"hardware":[42],"capacity.":[43],"existing":[45],"GPU":[46,207],"transformer-based":[48],"accelerators":[49],"cannot":[50],"efficiently":[51],"process":[52],"compressed":[53],"LLMs,":[54,151],"due":[55],"following":[58],"unresolved":[59],"challenges:":[60],"low":[61],"computational":[62],"efficiency,":[63],"underutilized":[64],"bandwidth,":[66],"large":[68],"compilation":[69,157,162],"This":[71],"paper":[72],"proposes":[73],"FlightLLM,":[74,87],"enabling":[75],"efficient":[76],"LLMs":[77,100,190],"inference":[78],"with":[79,126,141,208],"complete":[81],"mapping":[82],"flow":[83],"FPGAs.":[85],"In":[86],"we":[88,131,152],"highlight":[89],"an":[90,133],"innovative":[91],"solution":[92],"that":[93],"overhead":[98],"of":[99,201],"can":[101],"be":[102],"solved":[103],"by":[104],"utilizing":[105],"FPGA-specific":[106],"resources":[107],"(e.g.,":[108,185,191],"DSP48":[109],"heterogeneous":[111],"hierarchy).":[113],"We":[114],"propose":[115,132,153],"configurable":[117],"sparse":[118],"DSP":[119],"chain":[120],"support":[122],"different":[123],"sparsity":[124],"patterns":[125],"high":[127],"efficiency.":[129],"Second,":[130],"always-on-chip":[134],"decode":[135],"scheme":[136],"boost":[138],"bandwidth":[140],"mixed-precision":[142],"support.":[143],"Finally,":[144],"make":[146],"FlightLLM":[147,171,203],"available":[148],"for":[149],"real-world":[150],"length":[155],"adaptive":[156],"method":[158],"reduce":[160],"overhead.":[163],"Implemented":[164],"Xilinx":[167],"Alveo":[168],"U280":[169],"FPGA,":[170],"achieves":[172],"6.0\u00d7":[173],"higher":[174,210],"energy":[175],"1.8\u00d7":[178],"better":[179],"cost":[180],"against":[182],"commercial":[183],"GPUs":[184],"NVIDIA":[186,205],"V100S)":[187],"modern":[189],"LLaMA2-7B)":[192],"using":[193,212],"vLLM":[194],"SmoothQuant":[196],"under":[197],"batch":[199],"size":[200],"one.":[202],"beats":[204],"A100":[206],"1.2\u00d7":[209],"throughput":[211],"latest":[214],"Versal":[215],"VHK158":[216],"FPGA.":[217]},"counts_by_year":[{"year":2026,"cited_by_count":9},{"year":2025,"cited_by_count":72},{"year":2024,"cited_by_count":22}],"updated_date":"2026-04-16T08:26:57.006410","created_date":"2025-10-10T00:00:00"}