| Reducing activation recomputation in large transformer models VA Korthikanti, J Casper, S Lym, L McAfee, M Andersch, M Shoeybi, ... Proceedings of Machine Learning and Systems 5, 341-353, 2023 | 436 | 2023 |
| Prunetrain: fast neural network training by dynamic sparse model reconfiguration S Lym, E Choukse, S Zangeneh, W Wen, S Sanghavi, M Erez Proceedings of the International Conference for High Performance Computing …, 2019 | 130* | 2019 |
| Branchnet: A convolutional neural network to predict hard-to-predict branches S Zangeneh, S Pruett, S Lym, YN Patt 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture …, 2020 | 67 | 2020 |
| Near data acceleration with concurrent host access BY Cho, Y Kwon, S Lym, M Erez 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture …, 2020 | 65 | 2020 |
| DeLTA: GPU performance model for deep learning applications with in-depth memory system traffic analysis S Lym, D Lee, M O'Connor, N Chatterjee, M Erez 2019 IEEE international symposium on performance analysis of systems and …, 2019 | 60 | 2019 |
| Duo: Exposing on-chip redundancy to rank-level ecc for high reliability SL Gong, J Kim, S Lym, M Sullivan, H David, M Erez 2018 IEEE International Symposium on High Performance Computer Architecture …, 2018 | 56 | 2018 |
| All-inclusive ecc: Thorough end-to-end protection for reliable computer memory J Kim, M Sullivan, S Lym, M Erez ACM SIGARCH Computer Architecture News 44 (3), 622-633, 2016 | 46 | 2016 |
| Hamartia: A fast and accurate error injection framework CK Chang, S Lym, N Kelly, MB Sullivan, M Erez 2018 48th Annual IEEE/IFIP International Conference on Dependable Systems …, 2018 | 44 | 2018 |
| Evaluating and accelerating high-fidelity error injection for hpc CK Chang, S Lym, N Kelly, MB Sullivan, M Erez Sc18: International conference for high performance computing, networking …, 2018 | 40 | 2018 |
| Mini-batch Serialization: CNN Training with Inter-layer Data Reuse S Lym, A Behroozi, W Wen, G Li, Y Kwon, M Erez The Conference on Systems and Machine Learning (SysML), 2018 | 35 | 2018 |
| Nemotron-h: A family of accurate and efficient hybrid mamba-transformer models A Blakeman, A Basant, A Khattar, A Renduchintala, A Bercovich, A Ficek, ... arXiv preprint arXiv:2504.03624, 2025 | 31 | 2025 |
| FlexSA: Flexible systolic array architecture for efficient pruned DNN model training S Lym, M Erez arXiv preprint arXiv:2004.13027, 2020 | 31 | 2020 |
| Reducing activation recomputation in large transformer models, 2022 V Korthikanti, J Casper, S Lym, L McAfee, M Andersch, M Shoeybi, ... URL https://arxiv. org/abs/2205.05198, 0 | 28 | |
| ERUCA: Efficient DRAM resource utilization and resource conflict avoidance for memory system parallelism S Lym, H Ha, Y Kwon, C Chang, J Kim, M Erez 2018 IEEE International Symposium on High Performance Computer Architecture …, 2018 | 22 | 2018 |
| Write driver circuit, semiconductor apparatus using the same, and memory system SK LYM US Patent App. 13/720,739, 2014 | 14 | 2014 |
| Near data acceleration with concurrent host access. In 2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA) BY Cho, Y Kwon, S Lym, M Erez IEEE, 818ś831, 2020 | 10 | 2020 |
| Semiconductor device having memory chip stacks with TSV SK LYM US Patent 9,396,766, 2016 | 7 | 2016 |
| Current control apparatus and phase change memory having the same SK LYM, YJ Shin US Patent 8,526,226, 2013 | 7 | 2013 |
| Reference voltage generator SK LYM, YJ Shin US Patent 8,791,684, 2014 | 6 | 2014 |
| Input/output circuit and input/output device including the same S Lym US Patent 9,607,666, 2017 | 4 | 2017 |
Wei WenResearch Scientist, AI at MetaVerified email at fb.com
