Publications

Research on LLM agents, networked AI systems, model post-training, and network measurement.

CROP: Task Relevance via Counterfactuals for Selective On-Policy Distillation

Enhan Li, Junhao He, Hongyang Du

arXiv preprint arXiv:2608.13387 2026

Allocating token-level supervision using counterfactual task relevance for selective on-policy distillation.

PaperDOI

HACO: Hedged Agent Computing for Reliable LLM Systems

Enhan Li, Hongyang Du

Advances in Neural Information Processing Systems (NeurIPS, accepted) 2026

Reliable agent execution through adaptive hedge selection under changing network and service conditions.

Paper

JAUNT: Joint Alignment of User Intent and Network State for QoE-centric LLM Tool Routing

Enhan Li, Hongyang Du

arXiv preprint arXiv:2510.18550 2025

Aligning user intent with network conditions for QoE-aware LLM tool routing.

PaperDOI

NetMCP: Network-Aware Model Context Protocol Platform for LLM Capability Extension

Enhan Li, Hongyang Du, Kaibin Huang

arXiv preprint arXiv:2510.13467 2025

A network-aware platform and routing framework for extending LLMs with external tools and services.

PaperDOICode

Detecting and Defending Mechanism Against DDoS Attacks in Programmable Data Plane

Wen-Hao Wu, Lei-Lei Zhang, Heng Pan, En-Han Li, Jian-Er Zhou, Zhen-Yu Li

Journal of Software 2025

DDoS detection and real-time traffic filtering in programmable switches. (In Chinese)

PaperDOI

SAROS: A Self-Adaptive Routing Oblivious Sampling Method for Network-wide Heavy Hitter Detection

Enhan Li, Wenhao Wu, Zhaohua Wang, Zhenyu Li, Jianwei Niu

Proceedings of the 8th Asia-Pacific Workshop on Networking (APNet) 2024

Adaptive sampling for network-wide heavy hitter detection on programmable data planes.

PaperDOICode