Work

An Efficient Training Architecture for Nonlinear Softmax Function in Transformers

Year: 2025

Type: conference-paper

Source: IEEE International Symposium on Circuits and Systems proceedings

DOI: 10.1109/iscas56072.2025.11044130

Authors: Haikuo Shao, Zhongfeng Wang

Institution: Nanjing University

Language: English

Topic: Advanced Neural Network Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: closed

OpenAlex ID: W4411725376

FWCI: 1.563

Cites: 29

Cited by: 2

Related to: 10