5 papers
The Extrapolation Cliff in On-Policy Distillation of Near-Deterministic Structured Outputs
Xin Li, Hao Jiang, Annan Wang +2
On-policy distillation (OPD) is widely used for LLM post-training. When pushed with a reward-extrapolation coefficient lambda > 1, the student can lift past the teacher in domain,…
LLM Agent Communication Protocol (LACP) Requires Urgent Standardization: A Telecom-Inspired Protocol is Necessary
Xin Li, Mengbing Liu, Chau Yuen
This position paper argues that the field of LLM agents requires a unified, telecom-inspired communication protocol to ensure safety, interoperability, and scalability, especially…
Intra-DP: A High Performance Collaborative Inference System for Mobile Edge Computing
Zekai Sun, Xiuxian Guan, Zheng Lin +8
Deploying deep neural networks (DNNs) on resource-constrained mobile devices presents significant challenges, particularly in achieving real-time performance while simultaneously c…
A Novel Framework of Horizontal-Vertical Hybrid Federated Learning for EdgeIoT
Kai Li, Yilei Liang, Xin Yuan +4
This letter puts forth a new hybrid horizontal-vertical federated learning (HoVeFL) for mobile edge computing-enabled Internet of Things (EdgeIoT). In this framework, certain EdgeI…
Robust Gaussian Splatting SLAM by Leveraging Loop Closure
Zunjie Zhu, Youxu Fang, Xin Li +4
3D Gaussian Splatting algorithms excel in novel view rendering applications and have been adapted to extend the capabilities of traditional SLAM systems. However, current Gaussian…