Work

CADE: Context-aware dispatch engine for energy-efficient VLM inference on heterogeneous edge with discrete neural processing units

Year: 2026

Type: article

Source: Journal of Systems Architecture

DOI: 10.1016/j.sysarc.2026.104016

Authors: Seongjin Yun, Jeong-Si Kim

Institution: Electronics and Telecommunications Research Institute

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: closed

Funder: Institute for Information and Communications Technology Promotion

OpenAlex ID: W7214093136

FWCI:

Cites: 21

Cited by:

Related to: 10