Accelerating Cold Starts of On-Device LLMs via Multi-Source Inference-Aware Parameter Loading.
Shucheng Li, Zhenfeng Wang, Fei Zeng, Zhanxi Li, Peng Yang, Feng Lyu
Browse the full ICDCS paper archive.
Shucheng Li, Zhenfeng Wang, Fei Zeng, Zhanxi Li, Peng Yang, Feng Lyu
Browse the full ICDCS paper archive.