ELMS: Elasticized Large Language Models On Mobile Devices
원문 초록
On-device Large Language Models (LLMs) are revolutionizing mobile AI, enabling applications such as UI automation while addressing privacy concerns. Currently, the standard approach involves deploying a single, robust LLM as a universal solution for various applications, often referred to as LLM-as-a-Service (LLMaaS). However, this approach faces a significant system challenge: existing LLMs lack the flexibility to accommodate the diverse Service-Level Objectives (SLOs) regarding inference latency across different applications. To address this issue, we introduce ELMS, an on-device LLM service designed to provide elasticity in both the model and prompt dimensions of an LLMaaS. This system includes: A one-time neuron reordering technique, which utilizes the inherent permutation consistency within transformer models to create high-quality, elastic sub-models with minimal runtime switching costs. A dual-head compact language model, which efficiently refines prompts and coordinates the elastic adaptation between the model and the prompt. We have implemented this elastic on-device LLM service on several off-the-shelf (COTS) smartphones and evaluate ELMS using both standalone NLP/mobile-agent datasets and synthesized end-to-end traces. Across a range of SLOs, ELMS surpasses four strong baselines by up to 16.83% and 11.04% in absolute accuracy on average, with less than 1% Time-To-First-Token (TTFT) switching overhead, comparable memory usage, and fewer than 100 offline GPU hours.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
ShareGPT-4o-Image: Aligning Multimodal Models with GPT-4o-Level Image Generation
MNN-LLM: A Generic Inference Engine for Fast Large Language Model Deployment on Mobile Devices
Specification-based Security Evaluation of Real-World Agents Tool-Calling under Long-Horizon Context Tasks
Making Language Model a Hierarchical Classifier and Generator
Arctic Inference with Shift Parallelism: Fast and Efficient Open Source Inference System for Enterprise AI
InstructFLIP: Exploring Unified Vision-Language Model for Face Anti-spoofing
토론
이 논문과 연결된 포럼 글이 아직 없습니다.