{"id":"https://openalex.org/W4416429702","doi":"https://doi.org/10.1109/iccad66269.2025.11240748","title":"SAFF: Scalable Acceleration of GNN-based Machine Learning Force Fields using Tensor-Aware Hardware for Molecular Simulation","display_name":"SAFF: Scalable Acceleration of GNN-based Machine Learning Force Fields using Tensor-Aware Hardware for Molecular Simulation","publication_year":2025,"publication_date":"2025-10-26","ids":{"openalex":"https://openalex.org/W4416429702","doi":"https://doi.org/10.1109/iccad66269.2025.11240748"},"language":null,"primary_location":{"id":"doi:10.1109/iccad66269.2025.11240748","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad66269.2025.11240748","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","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/A5076372540","display_name":"Seohye Ha","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seohye Ha","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036617506","display_name":"Yunki Han","orcid":"https://orcid.org/0000-0003-0432-9324"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yunki Han","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100359405","display_name":"Tae-Hwan Kim","orcid":"https://orcid.org/0000-0002-9025-1350"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Taehwan Kim","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114228490","display_name":"Jiwan Kim","orcid":"https://orcid.org/0009-0009-5241-5631"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jiwan Kim","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100354901","display_name":"Junyoung Park","orcid":"https://orcid.org/0000-0001-8932-2820"},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Junyoung Park","raw_affiliation_strings":["Samsung Electronics,System LSI,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Electronics,System LSI,Republic of Korea","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Gunhee Park","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Gunhee Park","raw_affiliation_strings":["Samsung Electronics,System LSI,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Electronics,System LSI,Republic of Korea","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110193351","display_name":"Lee-Sup Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Lee-Sup Kim","raw_affiliation_strings":["Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea Advanced Institute of Science and Technology (KAIST),Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.45451729,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.9758999943733215,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.9758999943733215,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.0024999999441206455,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10044","display_name":"Protein Structure and Dynamics","score":0.0017999999690800905,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6625000238418579},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6575000286102295},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5551000237464905},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.4652000069618225},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4571000039577484},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.45190000534057617},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.41620001196861267},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.40709999203681946},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.4032999873161316},{"id":"https://openalex.org/keywords/computational-model","display_name":"Computational model","score":0.37720000743865967}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7285000085830688},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6625000238418579},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6575000286102295},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5551000237464905},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.48559999465942383},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.4652000069618225},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4571000039577484},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.45190000534057617},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.44839999079704285},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4431000053882599},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.41620001196861267},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.40709999203681946},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.4032999873161316},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.37720000743865967},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.37689998745918274},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.34540000557899475},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34060001373291016},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.3400999903678894},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3352999985218048},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.3337000012397766},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.32670000195503235},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3190000057220459},{"id":"https://openalex.org/C51255310","wikidata":"https://www.wikidata.org/wiki/Q1163016","display_name":"Tensor product","level":2,"score":0.29089999198913574},{"id":"https://openalex.org/C65232700","wikidata":"https://www.wikidata.org/wiki/Q5656403","display_name":"Hardware architecture","level":3,"score":0.29010000824928284},{"id":"https://openalex.org/C137981799","wikidata":"https://www.wikidata.org/wiki/Q1369184","display_name":"Reusability","level":3,"score":0.2856999933719635},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.2840000092983246},{"id":"https://openalex.org/C127964446","wikidata":"https://www.wikidata.org/wiki/Q1092142","display_name":"Computational resource","level":3,"score":0.28360000252723694},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.2815999984741211},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.27070000767707825},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C53442348","wikidata":"https://www.wikidata.org/wiki/Q745101","display_name":"Content-addressable memory","level":3,"score":0.26339998841285706},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2597000002861023},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2533999979496002},{"id":"https://openalex.org/C2778837361","wikidata":"https://www.wikidata.org/wiki/Q2450880","display_name":"Exascale computing","level":3,"score":0.2533999979496002}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccad66269.2025.11240748","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad66269.2025.11240748","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320315121","display_name":"Samsung Advanced Institute of Technology","ror":null},{"id":"https://openalex.org/F4320322202","display_name":"IC Design Education Center","ror":"https://ror.org/005v57z85"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W2054095206","https://openalex.org/W2138815064","https://openalex.org/W2725159389","https://openalex.org/W2761259030","https://openalex.org/W2807021761","https://openalex.org/W3006586535","https://openalex.org/W3159653252","https://openalex.org/W3176872900","https://openalex.org/W3189651386","https://openalex.org/W4225405705","https://openalex.org/W4287163433","https://openalex.org/W4391631778"],"related_works":[],"abstract_inverted_index":{"Machine":[0],"Learning":[1],"Force":[2],"Fields":[3],"(MLFFs)":[4],"have":[5,202],"emerged":[6],"as":[7,49],"a":[8,25,98,153,205,232],"key":[9],"technique":[10,157],"in":[11,66,126,235],"molecular":[12,20],"dynamics":[13],"(MD)":[14],"simulations,":[15],"enabling":[16],"accurate":[17],"prediction":[18],"of":[19,27,32,79,101,117,122,144,179,229],"energies":[21],"and":[22,57,75,119,130,169,182,190,231],"forces":[23],"with":[24,93],"level":[26],"precision":[28],"comparable":[29],"to":[30,45,90,136,158,194,239],"that":[31,139,211,221],"Density":[33],"Functional":[34],"Theory":[35],"(DFT),":[36],"while":[37,242],"significantly":[38],"reducing":[39,170],"computational":[40,81,103,146,196],"cost.":[41],"However,":[42],"applying":[43],"MLFFs":[44],"real-world":[46],"scenarios,":[47],"such":[48],"semiconductor":[50],"process":[51],"simulations":[52],"involving":[53],"massive":[54],"atomic":[55],"scales":[56],"long":[58],"simulation":[59],"times,":[60],"exposes":[61],"severe":[62],"performance":[63],"limitations,":[64],"resulting":[65],"current":[67],"MLFF":[68],"models":[69],"being":[70],"inadequate":[71],"for":[72,214],"practical":[73],"use":[74,143],"emphasizing":[76],"the":[77,102,115,120,141,187,222,244],"necessity":[78],"significant":[80],"acceleration.":[82],"Graph":[83],"Neural":[84],"Networks":[85],"(GNNs)":[86],"are":[87,112],"frequently":[88],"utilized":[89],"construct":[91],"MLFFs,":[92],"tensor":[94,207],"product":[95,208],"operations":[96,106,193],"constituting":[97],"substantial":[99],"portion":[100],"workload.":[104],"These":[105],"involve":[107],"high-dimensional":[108],"tensors":[109],"whose":[110],"sizes":[111],"determined":[113],"by":[114,185],"number":[116],"channels":[118],"order":[121],"rotation.":[123],"This":[124,134],"results":[125,219],"frequent":[127],"memory":[128,137,172],"accesses":[129],"low":[131],"data":[132,160,167],"reuse.":[133],"leads":[135],"bottlenecks":[138],"limit":[140],"efficient":[142],"GPU":[145],"resources.":[147],"In":[148],"this":[149,184],"work,":[150],"we":[151,175,201],"propose":[152],"tensor-based":[154],"function":[155],"fusion":[156],"improve":[159],"continuity":[161],"between":[162],"functions,":[163],"thereby":[164],"increasing":[165],"on-chip":[166],"reuse":[168],"off-chip":[171],"access.":[173],"Furthermore,":[174],"identified":[176],"inefficient":[177],"utilization":[178],"hardware":[180,209],"resources":[181],"addressed":[183],"restructuring":[186],"computation":[188],"flow":[189],"removing":[191],"redundant":[192],"enhance":[195],"performance.":[197],"To":[198],"optimize":[199],"performance,":[200],"also":[203],"designed":[204],"dedicated":[206],"architecture":[210],"is":[212],"optimized":[213],"GNN-based":[215],"MLFFs.":[216],"The":[217],"experimental":[218],"demonstrate":[220],"proposed":[223],"system":[224],"achieves":[225],"an":[226],"average":[227],"speedup":[228],"6.26\u00d7":[230],"424\u00d7":[233],"improvement":[234],"energy":[236],"efficiency":[237],"compared":[238],"GPU-based":[240],"execution,":[241],"maintaining":[243],"original":[245],"model":[246],"accuracy.":[247]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-20T00:00:00"}