Work

A 4.21 TFLOPS/W Memory-Efficient LLM Inference Accelerator with Bit-Layered Non-Uniform Quantization

Year: 2025

Type: conference-paper

Source: IEEE International Symposium on Circuits and Systems proceedings

DOI: 10.1109/iscas56072.2025.11043714

Authors: Byeongcheol Kim, Sangjin Kim, Sangwoo Ha, Soyeon Um, Kyomin Sohn, Hoi-Jun Yoo

Institutions: Korea Advanced Institute of Science and Technology, Samsung (South Korea), Samsung Electronics (South Korea)

Language: English

Topic: Advanced Neural Network Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: closed

Funders: IC Design Education Center, Samsung

OpenAlex ID: W4411725788

FWCI:

Cites: 14

Cited by:

Related to: 10