Lab directory
Kim Yiseob Lab
Kim Yiseob Lab
KAIST · Electrical Engineering
Information sourcenot verified yet
Professor
Prof. 김이섭
Professor's page · publicationsAbout
KAIST 전기및전자공학부 김이섭(Lee-Sup Kim) ICT 석좌교수의 MVLSI(Multimedia VLSI) 연구실이다. 에너지 효율적인 딥러닝 하드웨어 아키텍처와 신경망용 인-메모리 컴퓨팅의 VLSI 구현을 연구한다. CNN·RNN 등 딥러닝 알고리즘 기반 지능형 하드웨어 설계가 주요 대상이다.
Research keywords
Energy-efficient Deep-learning HardwareNeural Network VLSIIn-memory Computing
Selected publications
- AToM: Adaptive Token Merging for Efficient Acceleration of Vision TransformerIEEE Transactions on Computers · 2025
- EOD: Enabling Low Latency GNN Inference via Near-Memory Concatenate AggregationISCA (International Symposium on Computer Architecture) · 2025
- ADC-Free ReRAM-Based In-Situ Accelerator for Energy-Efficient Binary Neural NetworksIEEE Transactions on Computers · 2024
- Token-Picker: Accelerating Attention in Text Generation with Minimized Memory TransferDAC (Design Automation Conference) · 2024
- MGen: A Framework for Energy-Efficient In-ReRAM Acceleration of Multi-Task BERTIEEE Transactions on Computers · 2023
- Accelerating Deep Reinforcement Learning via Phase-Level Parallelism for RoboticsIEEE Computer Architecture Letters · 2024
Other labs in this department
- Kim Dongjoon LabRecruiting
Dongjun Kim
Computer architecture, interconnection networks, mobile systems, hardware security
- Kim Moonchul LabRecruiting
Munchurl Kim
Video coding, pattern recognition, visual communications
- Kim Minjun LabRecruiting
Min Jun Kim
Physical robot interaction, collaborative robots, aerial manipulation
- Kim Seongmin LabRecruiting
Song Min Kim
Wireless systems and networks, mobile computing, IoT
- Kim Yonghoon LabRecruiting
Yong-Hoon Kim
Atomic-scale device simulations, high-performance/machine-learning scientific computing
- Kim Junmo LabRecruiting
Junmo Kim
Statistical signal processing, image processing, computer vision, information theory