{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,20]],"date-time":"2025-10-20T18:57:26Z","timestamp":1760986646324,"version":"build-2065373602"},"reference-count":4,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada (NSERC) Alliance Grants","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Mitacs Accelerate Program"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Embedded Syst. Lett."],"published-print":{"date-parts":[[2025,10]]},"DOI":"10.1109\/les.2025.3601057","type":"journal-article","created":{"date-parts":[[2025,10,16]],"date-time":"2025-10-16T17:37:45Z","timestamp":1760636265000},"page":"317-320","source":"Crossref","is-referenced-by-count":0,"title":["Fused Tensor Core: A Hardware\u2013Software Co-Design for Efficient Execution of Attentions on GPUs"],"prefix":"10.1109","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-3555-9957","authenticated-orcid":false,"given":"Reza","family":"Jahadi","sequence":"first","affiliation":[{"name":"Electrical and Computer Engineering Department, Lakehead University, Thunder Bay, ON, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Phil","family":"Munz","sequence":"additional","affiliation":[{"name":"TrojAI, Saint John, NB, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1662-5334","authenticated-orcid":false,"given":"Ehsan","family":"Atoofian","sequence":"additional","affiliation":[{"name":"Electrical and Computer Engineering Department, Lakehead University, Thunder Bay, ON, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS.2019.00016"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD57390.2023.10323651"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/FPL60245.2023.00012"},{"key":"ref4","article-title":"FlashAttention: Fast and memory-efficient exact attention with IO-awareness","author":"Dao","year":"2022","journal-title":"arXiv:2205.14135"}],"container-title":["IEEE Embedded Systems Letters"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/4563995\/11205899\/11205900.pdf?arnumber=11205900","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,20]],"date-time":"2025-10-20T17:49:55Z","timestamp":1760982595000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11205900\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10]]},"references-count":4,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/les.2025.3601057","relation":{},"ISSN":["1943-0663","1943-0671"],"issn-type":[{"type":"print","value":"1943-0663"},{"type":"electronic","value":"1943-0671"}],"subject":[],"published":{"date-parts":[[2025,10]]}}}