AI NewsWaveLG Uplus Has Two AI Token Optimization Papers Accepted at NeurIPS Workshop
Key points
- Two AI token optimization papers accepted at NeurIPS workshop
- Smartphone AI processing speed up about 4.5 times
- Battery consumption down about 77%
LG Uplus said on the 29th that two papers on AI token optimization research were accepted at the On-Device Intelligence workshop of NeurIPS 2026. According to iNews24 and Digital Today, NeurIPS is regarded as one of the world's three major AI conferences, along with ICML and ICLR.
IT Chosun, EBN and Korea NGO Times reported that the research increased AI processing speed on smartphones by about 4.5 times and cut battery consumption by about 77%. IT Chosun explained that the technology allows AI to run efficiently even with a smartphone's limited memory and power. Digital Daily described the research as an AI token diet.
Recently, there has been a steady stream of AI-related news from LG Uplus. Digital Today and iNews24 reported that on the 28th, LG Uplus ranked first in the IPTV category of the 2026 AI Transformation (AX) Quality Satisfaction Index, organized by the Korean Standards Association. Digital Today explained that the index evaluates the quality of AI-applied products and services from the perspective of the AI experience customers feel. According to Techworld News, the Ministry of Education said it will sign a memorandum of understanding with LG Uplus and Theori Korea on the 31st at LG Science Park in Seoul to support the operation of a gifted education center for AI and information security.
Glossary A token is the basic unit an AI language model uses to process text. On-device AI refers to a method in which AI computation runs directly inside a device such as a smartphone, without going through a server.