Mobile AI

Mobile AI focuses on deploying artificial intelligence, particularly large language and multimodal models, directly onto mobile devices to enhance privacy, personalization, and offline functionality. Current research emphasizes optimizing model efficiency through techniques like quantization and developing novel architectures (e.g., elastic LLMs) to meet diverse application needs and resource constraints, often benchmarking performance on various mobile platforms and NPUs. This field is significant for its potential to improve accessibility of AI-powered applications in resource-limited settings and to address concerns about data privacy and latency in mobile computing.

Papers

November 7, 2022