MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases (arXiv:2402.14905)

arXiv:2402.14905. MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases. By Zechun Liu et al. (Meta). Submitted 2024-02-22. Optimizes sub-billion parameter LLMs for on-device inference with efficient architecture design. 2026-10-10: Fetched https://arxiv.org/abs/2402.14905. Title: 'MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases'. Peer-reviewed venue confirmed: ICML 2024. Abstract confirms architecture design optimizing sub-billion parameter LLMs for on-device inference.