Conference

2025

• TrioN: 3N0C RPU Cell with Minimal Power Consumption, Compact Footprint for Cross-point Array-based Analog AI Accelerator

• First Experimental Demonstration of Single-crystalline WO3-x Nanowire-based ECRAMs

• Operation Methodologies and Device Specifications for Analog AI Training Accelerator with Resistive Cross-point Arrays

• Symmetry Improvement in IGZO-based TrioN (3N0C) Cells for AI Computation

• Revealing the Switching Mechanism of ECRAM through an Independently-Contacted WO3-x Double Layer ECRAM

• Design Optimization of Capacitor-Based Synaptic Cells for Efficient Analog Neural Network Training

2024

• Excellent Synaptic Characteristics and Half-bias Selectivity in Vertical Short-channel ECRAM and Selector-free 4k Cross-point Array Demonstration

• Investigation of Contact Resistance between WOx Channel and Metal Electrodes in Electrochemical Ramdom-Access Memory

• Design Strategies of 3T1C Synapse for High-efficiency Analog Neural Network Training

• Revealing Oxygen Donor Level in Tungsten Oxide Films for Neuromorphic Applications with Parallel Dipole-Line Hall System

• Investigation of Contact Resistance between WOx Channel and Metal Electrodes in Electrochemical Ramdom-Access Memory

2023

• An ECRAM-Based Analog Compute-in-Memory Neuromorphic System with High-Precision Current Readout 

• Reservoir Optimization in Oxygen-Based Electrochemical Random-Access Memory for Improved Synaptic Characterstics

• 시냅스 소자의 Retention 특성이 Tiki-Taka 알고리즘 기반 뉴럴 네트워크 학습의 정확도에 미치는 영향에 관한 연구

• High Performance Neural Network Inference with a 3D NAND-like Synaptic Weight Storage

• A Study on Metal-oxide Based ECRAM Device Characteristics of Selector-free Cross-point Array for DNN