@inproceedings{oh2026qflash,author={Oh, Sehyeon and Kwon, Yongin and Lee, Jemin},title={QFlash: Bridging Quantization and Memory Efficiency in Vision Transformer Attention},booktitle={International Joint Conference on Artificial Intelligence (IJCAI)},year={2026},month=aug,note={To appear},}
@article{kwon2025luthier,author={Kwon, Yongin and Cha, JooHyoung and Oh, Sehyeon and Yu, Misun and Park, Jeman and Lee, Jemin},title={Luthier: Bridging Auto-Tuning and Vendor Libraries for Efficient Deep Learning Inference},journal={ACM Transactions on Embedded Computing Systems},volume={24},number={5s},pages={92:1--92:23},year={2025},month=sep,publisher={Association for Computing Machinery},doi={10.1145/3759916},url={https://dl.acm.org/doi/10.1145/3759916},note={Presented at ESWEEK CASES 2025 (journal track)},}
@inproceedings{oh2025iflashattention,author={Oh, Sehyeon and Kwon, Yongin and Lee, Jemin},title={Work-in-Progress: I-FlashAttention: Fully Integer Fused Attention for Efficient Vision Transformers},booktitle={International Conference on Compilers, Architecture, and Synthesis for Embedded Systems (CASES)},pages={25--26},year={2025},month=sep,publisher={Association for Computing Machinery},doi={10.1145/3742872.3757072},}
@inproceedings{lee2025triplannet,author={Lee, Choonghan and Harris, Leah and Oh, Sehyun and Cha, JooHyoung and Lee, Jemin and Kwon, Yongin and Bae, Andrew Jang-ho},title={TriPlanNet: Triangle Path Planning Network for A Variable Truss Robot with Deep Learning},booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops},pages={2497--2506},year={2025},month=oct,doi={10.1109/ICCVW69036.2025.00260},note={13th International Workshop on Assistive Computer Vision and Robotics (ACVR)},}
@article{oh2025data,author={Oh, Sehyeon and Kwon, Yongin and Lee, Jemin},title={Data Transmission Optimization for NPU System Scalability},journal={IEMEK Journal of Embedded Systems and Applications},volume={20},number={3},pages={125--130},year={2025},month=jun,publisher={Institute of Embedded Engineering of Korea},doi={10.14372/IEMEK.2025.20.3.125},}
@article{oh2025optimizing,author={Oh, Sehyeon and Kwon, Yongin and Lee, Jemin},title={Optimizing Real-Time Object Detection in a Multi-Neural Processing Unit System},journal={Sensors},volume={25},number={5},pages={1376},year={2025},month=mar,publisher={MDPI},doi={10.3390/s25051376},url={https://www.mdpi.com/1424-8220/25/5/1376},note={Invited extension of our AICompS 2024 Distinguished Paper},}
2024
AICompS
Optimizing Real-Time Object Detection in a Multi NPU System with Double Buffering and Queue-Based Processing
Distinguished Paper Award at AICompS 2024, which led to an invited journal extension in MDPI Sensors.
@inproceedings{oh2024aicomps,author={Oh, Sehyeon and Kwon, Yongin and Lee, Jemin},title={Optimizing Real-Time Object Detection in a Multi NPU System with Double Buffering and Queue-Based Processing},booktitle={The 1st International Conference on Artificial Intelligence Computing and Systems (AICompS)},year={2024},month=dec,}
Distinguished Paper Award at the IEMEK Autumn Conference 2024.
@inproceedings{oh2024pcie,author={Oh, Sehyeon and Kwon, Yongin and Lee, Jemin},title={PCIe-Based Multi-NPU Data Transmission Optimization},booktitle={IEMEK Autumn Conference},year={2024},month=nov,}