{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T07:14:01Z","timestamp":1772694841624,"version":"3.50.1"},"reference-count":68,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T00:00:00Z","timestamp":1769817600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T00:00:00Z","timestamp":1769817600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,31]]},"DOI":"10.1109\/hpca68181.2026.11408495","type":"proceedings-article","created":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T20:47:22Z","timestamp":1772657242000},"page":"1-14","source":"Crossref","is-referenced-by-count":0,"title":["FlashFuser: Expanding the Scale of Kernel Fusion for Compute-Intensive Operators via Inter-Core Connection"],"prefix":"10.1109","author":[{"given":"Ziyu","family":"Huang","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yangjie","family":"Zhou","sequence":"additional","affiliation":[{"name":"National University of Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zihan","family":"Liu","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xinhao","family":"Luo","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yijia","family":"Diao","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Minyi","family":"Guo","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jidong","family":"Zhai","sequence":"additional","affiliation":[{"name":"Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yu","family":"Feng","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chen","family":"Zhang","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anbang","family":"Wu","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingwen","family":"Leng","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC55821.2022.9926299"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651330"},{"key":"ref3","first-page":"578","article-title":"{TVM}: An automated {End-to-End} optimizing compiler for deep learning","volume-title":"13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18)","author":"Chen","year":"2018"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/HCS55958.2022.9895592"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/HCS49909.2020.9220622"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3653019"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3576933"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3617232.3624864"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651351"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589038"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640423"},{"key":"ref12","article-title":"Waferllm: Large language model inference at wafer scale","author":"He","year":"2025","journal-title":"arXiv preprint"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-32041-5_3"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA61900.2025.00086"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3410463.3414623"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00047"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359630"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO61859.2024.00070"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3575747"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/HCS52781.2021.9567075"},{"key":"ref21","article-title":"Bert: a review of applications in natural language processing and understanding","author":"Koroteev","year":"2021","journal-title":"arXiv preprint"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3694715.3695955"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA61900.2025.00112"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640360"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC62836.2024.10938416"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS57955.2024.00064"},{"key":"ref27","first-page":"2508","article-title":"Clusterfusion: Expanding operator fusion scope for llm inference via cluster-level collective primitive","author":"Luo","year":"2025","journal-title":"arXiv eprints"},{"key":"ref28","first-page":"881","article-title":"Rammer: Enabling holistic deep learning compiler optimizations with \\{rTasks\\}","volume-title":"14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20)","author":"Ma","year":"2020"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/SCW63240.2024.00101"},{"key":"ref30","article-title":"Comet: A framework for modeling compound operation dataflows with explicit collectives","author":"Negi","year":"2025","journal-title":"arXiv preprint"},{"key":"ref31","volume-title":"TensorRT: A High-Performance Deep Learning Inference SDK","year":"2025"},{"key":"ref32","volume-title":"Nvidia cublas. NVIDIA Corporation","year":"2025"},{"key":"ref33","article-title":"Cuda c++ programming guide 4.6.10. distributed shared memory","volume-title":"NVIDIA Corporation","year":"2025"},{"key":"ref34","volume-title":"CUDA C++ Programming Guide","year":"2025"},{"key":"ref35","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3297858.3304059"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/2499370.2462176"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2023.3307796"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3211346.3211348"},{"key":"ref40","first-page":"701","article-title":"Welder: Scheduling deep learning memory access via tilegraph","volume-title":"17th USENIX Symposium on Operating Systems Design and Implementation (OSDI 23)","author":"Shi","year":"2023"},{"key":"ref41","author":"Thakkar","year":"2023","journal-title":"CUTLASS"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3315508.3329973"},{"key":"ref43","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref44","article-title":"Tilelang: A composable tiled programming model for ai systems","author":"Wang","year":"2025","journal-title":"arXiv preprint"},{"key":"ref45","first-page":"307","article-title":"Ladder: Enabling efficient {Low-Precision} deep learning computing through hardware-aware tensor transformation","volume-title":"18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24)","author":"Wang","year":"2024"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2025.3549621"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0153"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CGO51591.2021.9370330"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3624062.3626079"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1145\/3617232.3624858"},{"key":"ref51","first-page":"204","article-title":"Bolt: Bridging the gap between auto-tuners and hardware-native performance","volume-title":"Proceedings of Machine Learning and Systems","volume":"4","author":"Xing","year":"2022"},{"key":"ref52","article-title":"vtensor: Flexible virtual tensor management for efficient 11 m serving","author":"Xu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3575737"},{"key":"ref54","first-page":"681","article-title":"Cocktailer: Analyzing and optimizing dynamic control flow in deep learning","volume-title":"17th USENIX Symposium on Operating Systems Design and Implementation (OSDI 23)","author":"Zhang","year":"2023"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/SC41406.2024.00040"},{"key":"ref56","first-page":"863","article-title":"Ansor: Generating \\{High-Performance\\} tensor programs for deep learning","volume-title":"14th USENIX symposium on operating systems design and implementation (OSDI 20)","author":"Zheng","year":"2020"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2000"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527440"},{"key":"ref59","doi-asserted-by":"crossref","first-page":"1271","DOI":"10.1145\/3613424.3623792","article-title":"Tileflow: A framework for modeling fusion dataflow via tree-based analysis","volume-title":"Proceedings of the 56th Annual IEEE\/ACM International Symposium on Microarchitecture","author":"Zheng","year":"2023"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA56546.2023.10071018"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378508"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507723"},{"key":"ref63","article-title":"Fusionstitching: boosting memory intensive computations for deep learning workloads","author":"Zheng","year":"2020","journal-title":"arXiv preprint"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3575723"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1145\/3676642.3736121"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1145\/3712285.3759779"},{"key":"ref67","first-page":"233","article-title":"{ROLLER}: Fast and efficient tensor compilation for deep learning","volume-title":"16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22)","author":"Zhu","year":"2022"},{"key":"ref68","first-page":"989","article-title":"{MonoNN}: Enabling a new monolithic optimization space for neural network inference tasks on modern {GPU-Centric} architectures","volume-title":"18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24)","author":"Zhuang","year":"2024"}],"event":{"name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","location":"Sydney, Australia","start":{"date-parts":[[2026,1,31]]},"end":{"date-parts":[[2026,2,4]]}},"container-title":["2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11408404\/11408433\/11408495.pdf?arnumber=11408495","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T06:35:15Z","timestamp":1772692515000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11408495\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,31]]},"references-count":68,"URL":"https:\/\/doi.org\/10.1109\/hpca68181.2026.11408495","relation":{},"subject":[],"published":{"date-parts":[[2026,1,31]]}}}